EDataset manages a dataset of images. More...

Inheritance diagram for eVision.EasyDeepLearning.EDataset:

[legend]

Collaboration diagram for eVision.EasyDeepLearning.EDataset:

Public Member Functions
	EDataset ()
	Constructs a EDataset object.
	EDataset (eVision.EasyDeepLearning.EDataset other)
	Copy constructor of a EDataset object.
void	Load (string path)
	Loads a classification dataset from disk.
void	Load (eVision.ESerializer serializer)
	Loads a classification dataset from disk.
void	Save (string path)
	Saves a classification dataset to disk, containing the file paths to the images in the dataset and their associated label.
void	Save (eVision.ESerializer serializer)
	Saves a classification dataset to disk, containing the file paths to the images in the dataset and their associated label.
void	SaveWithRelativePath (string path)
	Saves the dataset with image paths relative to the directory in which the dataset file is saved.
void	Export (string directory, eVision.EImageFileType fileType, int quality)
	Exports the dataset and its images to the given directory.
void	Export (string directory, eVision.EImageFileType fileType)
	Exports the dataset and its images to the given directory.
void	Export (string directory)
	Exports the dataset and its images to the given directory.
void	Export (string directory, bool keepFilename, eVision.EImageFileType fileType, int quality)
	Exports the dataset and its images to the given directory.
void	Export (string directory, bool keepFilename, eVision.EImageFileType fileType)
	Exports the dataset and its images to the given directory.
void	Export (string directory, bool keepFilename)
	Exports the dataset and its images to the given directory.
void	Export (string directory, int width, int height, int channels, eVision.EImageFileType fileType, int quality)
	Exports the dataset and its images to the given directory.
void	Export (string directory, int width, int height, int channels, eVision.EImageFileType fileType)
	Exports the dataset and its images to the given directory.
void	Export (string directory, int width, int height, int channels)
	Exports the dataset and its images to the given directory.
void	Export (string directory, int width, int height, int channels, bool keepFilename, eVision.EImageFileType fileType, int quality)
	Exports the dataset and its images to the given directory.
void	Export (string directory, int width, int height, int channels, bool keepFilename, eVision.EImageFileType fileType)
	Exports the dataset and its images to the given directory.
void	Export (string directory, int width, int height, int channels, bool keepFilename)
	Exports the dataset and its images to the given directory.
virtual void	Clear ()
	Clears the dataset.
eVision.EasyDeepLearning.EImageAnnotationFormat	GetAvailableImageAnnotationFormat (string imageFile)
	Image annotation file formats that are available next to the image file.
eVision.EasyDeepLearning.EImageAnnotationFormat	GetAvailableImageAnnotationFormat (int imgId)
	Image annotation file formats that are available next to the image file.
void	ImportYOLOTXTAnnotations (int imgId, string labelFile)
	Imports YOLO TXT annotations (for EasyLocate Bounding Box).
void	ImportYOLOTXTAnnotations (string labelFile)
	Imports YOLO TXT annotations (for EasyLocate Bounding Box).
void	ImportPascalVOCXMLAnnotations (int imgId)
	Imports Pascal VOC XML annotations (for EasyLocate Bounding Box).
void	ImportPascalVOCXMLAnnotations ()
	Imports Pascal VOC XML annotations (for EasyLocate Bounding Box).
int	AddImage (string imagePath)
	Adds an image to the dataset.
int	AddImage (string imagePath, string label)
	Adds an image to the dataset.
int	AddImage (string imagePath, int originX, int originY, int width, int height, eVision.ERegion mask)
	Adds an image to the dataset.
int	AddImage (string imagePath, int originX, int originY, int width, int height, eVision.ERegion mask, string label)
	Adds an image to the dataset.
int	AddImage (eVision.EBaseROI img)
	Adds an image to the dataset.
int	AddImage (eVision.EBaseROI img, string label)
	Adds an image to the dataset.
int	AddImage (eVision.EBaseROI img, int originX, int originY, int width, int height, eVision.ERegion mask)
	Adds an image to the dataset.
int	AddImage (eVision.EBaseROI img, int originX, int originY, int width, int height, eVision.ERegion mask, string label)
	Adds an image to the dataset.
int	AddImages (string filter)
	Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.
int	AddImages (string filter, string label)
	Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.
int	AddImages (string filter, int originX, int originY, int width, int height, eVision.ERegion mask)
	Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.
int	AddImages (string filter, int originX, int originY, int width, int height, eVision.ERegion mask, string label)
	Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.
int	AddImagesFromCSV (string csvFile, int imageColumn, int labelColumn, int skipLines, string separator)
	Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.
int	AddImagesFromCSV (string csvFile, int imageColumn, int labelColumn, int skipLines)
	Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.
int	AddImagesFromCSV (string csvFile, int imageColumn, int labelColumn)
	Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.
int	AddImagesFromCSV (string csvFile, int imageColumn)
	Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.
int	AddImagesFromCSV (string csvFile)
	Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.
void	RemoveImage (int imgId)
	Removes the image at the given index. All annotations (label, segmentation, objects) will be lost.
int	GetNumImagesForFile (string filePath)
	Get the number of images in the dataset that are associated with the given image file.
eVision.EBaseROI	GetImageCopy (int index)
	Gets a copy of the i-th image of the dataset.
void	GetImageCopy (int index, eVision.EBaseROI img)
	Gets a copy of the i-th image of the dataset.
string	GetImagePath (int index)
	Gets the path of the i-th image of the dataset. If the image was not given using a path, the method will throw an exception.
bool	IsImageFile (int index)
	Whether the image is stored as a file path towards an image.
bool	IsEmbeddedImage (int index)
	Whether the image is embedded into the dataset and is not a reference towards an image file.
string	GetImageLabel (int index)
	Gets the label of the i-th image of the dataset. An exception is thrown if the image has no classification label.
void	SetImageLabel (int index, string label)
	Sets the label of images in the dataset.
void	SetImageLabel (string filter, string label)
	Sets the label of images in the dataset.
void	UnsetImageLabel (int index)
	Unset the label of the given image of the dataset. After this operation, the given image will have no classification labels.
bool	HasLabel (int index)
	Whether an image has a classification label.
uint[]	GetImagesIndexesWithLabel (string label)
	Gets a list of index corresponding to the images associated with the given label.
uint[]	GetImagesIndexesWithLabel (int labelIndex)
	Gets a list of index corresponding to the images associated with the given label.
void	AddLabel (string label)
	Adds a label to the dataset.
string	GetLabel (int i)
	Gets the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1).
void	SetLabel (int index, string label)
	Sets the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1). This operation does not add a new label to the dataset but simply renames an existing label.
float	GetLabelWeight (int index)
	Gets the weight associated to the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1).
void	SetLabelWeight (int index, float weight)
	Sets the weight associated to the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1). This operation does not add a new label.
void	SetLabelColor (int index, eVision.ERGBColor color)
	Sets the label color for a classification label.
eVision.ERGBColor	GetLabelColor (int index)
	Gets the label color for a classification label.
void	SetLabelOpacity (int index, byte opacity)
	Sets the label opacity for a classification label.
byte	GetLabelOpacity (int index)
	Gets the label opacity for a classification label.
void	RemoveLabel (string label)
	Removes a label from the dataset.
void	RemoveLabel (int labelId)
	Removes a label from the dataset.
void	SetSegmentationMap (int imageIndex, eVision.EROIBW16 segmentationMap)
	Sets the segmentation map of an image. The segmentation map is a 16-bit EROIBW16 image where the value of each pixel is equal to the corresponding segmentation label index (see EDataset.GetSegmentationLabel).
eVision.EImageBW16	GetSegmentationMap (int imageIndex)
	Gets the segmentation map of an image. The segmentation map is a 16-bit EROIBW16 image where the value of each pixel is equal to the corresponding segmentation label index (see EDataset.GetSegmentationLabel).
void	UnsetSegmentation (int imageIndex)
	Unsets the segmentation of an image. After this operation, the image will have no segmentation.
void	ResetSegmentation (int imageIndex)
	Resets the segmentation of an image by setting all pixels to background.
void	AddRegionToSegment (int imageIndex, int segmentationLabelIndex, eVision.ERegion region)
	Adds a region (see ERegion) to the given segment of an image. If, before this call, the image had no segmentation, the pixels not in the given region will be set to the background segmentation label.
void	AddRegionToSegment (int imageIndex, string label, eVision.ERegion region)
	Adds a region (see ERegion) to the given segment of an image. If, before this call, the image had no segmentation, the pixels not in the given region will be set to the background segmentation label.
eVision.ERegion	GetRegionForSegment (int imageIndex, int segmentationLabelIndex)
	Gets a region corresponding to the pixels of the given segment.
eVision.ERegion	GetRegionForSegment (int imageIndex, string segmentationLabel)
	Gets a region corresponding to the pixels of the given segment.
int	GetNumImagesForSegmentationLabel (int labelId)
	Number of images containing the given segmentation label.
int	GetNumImagesForSegmentationLabel (string label)
	Number of images containing the given segmentation label.
ulong	GetNumPixelsForSegmentationLabel (int labelId)
	Number of pixels assigned to the given segmentation label.
ulong	GetNumPixelsForSegmentationLabel (string label)
	Number of pixels assigned to the given segmentation label.
bool	HasSegmentation (int imageId)
	Whether the image has a segmentation.
bool	HasForegroundSegments (int imageId)
	Whether the image segmentation contains segments that are not background. If the image has no segmentation, this method throws an exception.
int	GetNumSegmentedBlobs (int imageId)
	Number of segmented blobs in the image for all non-background labels or for a specific label.
int	GetNumSegmentedBlobs (int imageId, string label)
	Number of segmented blobs in the image for all non-background labels or for a specific label.
int	GetNumSegmentedBlobs (int imageId, int labelId)
	Number of segmented blobs in the image for all non-background labels or for a specific label.
string	GetSegmentationLabel (int index)
	Gets the segmentation label.
void	SetSegmentationLabel (int index, string label)
	Sets the segmentation labels.
int	AddSegmentationLabel (string label, float labelWeight)
	Adds a segmentation label. The index of the new segmentation labels is returned by the method.
int	AddSegmentationLabel (string label)
	Adds a segmentation label. The index of the new segmentation labels is returned by the method.
void	RemoveSegmentationLabel (string label)
	Remove a segmentation label.
void	RemoveSegmentationLabel (int index)
	Remove a segmentation label.
float	GetSegmentationLabelWeight (int index)
	Gets the segmentation label weights.
float	GetSegmentationLabelWeight (string label)
	Gets the segmentation label weights.
void	SetSegmentationLabelWeight (int index, float weight)
	Sets the segmentation label weight.
void	SetSegmentationLabelWeight (string label, float weight)
	Sets the segmentation label weight.
void	SetSegmentationLabelColor (int index, eVision.ERGBColor color)
	Sets the label color for a segmentation label.
eVision.ERGBColor	GetSegmentationLabelColor (int index)
	Gets the label color for a segmentation label.
void	SetSegmentationLabelOpacity (int index, byte opacity)
	Sets the label opacity for a segmentation label.
byte	GetSegmentationLabelOpacity (int index)
	Gets the label opacity for a segmentation label.
void	ResetImageObjectLabeling (int imageIndex)
	Resets the object labeling for the specified image. This sets the image as having been labeled for object detection with no object present in the image.
void	UnsetImageObjectLabeling (int imageIndex)
	Unsets the object labeling for the specified image. This sets the image as not having been labeled for object detection. This image won't be use for training with a ELocator tool.
eVision.EasyDeepLearning.ELocatorObject[]	GetImageObjects (int imageIndex)
	Gets all the objects of the specified image.
int	GetImageNumObjects (int imageIndex)
	Number of objects for the specified image.
eVision.EasyDeepLearning.ELocatorObject	GetImageObject (int imageIndex, int objectIndex)
	Gets the specified object for the given image.
void	SetImageObject (int imageIndex, int objectIndex, eVision.EasyDeepLearning.ELocatorObject obj)
	Sets the specified object for the given image.
void	SetImageObject (int imageIndex, int objectIndex, string label)
	Sets the specified object for the given image.
void	SetImageObject (int imageIndex, int objectIndex, eVision.ERectangleRegion region)
	Sets the specified object for the given image.
void	AddImageObject (int imageIndex, eVision.EasyDeepLearning.ELocatorObject obj)
	Adds an object to the image.
void	AddImageObject (int imageIndex, string label, eVision.ERectangleRegion box)
	Adds an object to the image.
void	RemoveImageObject (int imageIndex, int objectIndex)
	Removes an object from an image.
string	GetObjectLabel (int labelIndex)
	Object label.
void	SetObjectLabel (int labelIndex, string newLabel)
	Sets an object label.
void	SetObjectLabel (string oldLabel, string newLabel)
	Sets an object label.
void	AddObjectLabel (string label, float weight)
	Adds an object label.
void	AddObjectLabel (string label)
	Adds an object label.
void	RemoveObjectLabel (int labelIndex)
	Removes an object label.
void	RemoveObjectLabel (string label)
	Removes an object label.
void	SetObjectLabelWeight (int labelIndex, float weight)
	Sets the weight of an object label.
void	SetObjectLabelWeight (string label, float weight)
	Sets the weight of an object label.
float	GetObjectLabelWeight (int labelIndex)
	Gets the weight of an object label.
float	GetObjectLabelWeight (string label)
	Gets the weight of an object label.
void	SetObjectLabelColor (int index, eVision.ERGBColor color)
	Sets the label color for an object label.
eVision.ERGBColor	GetObjectLabelColor (int index)
	Gets the label color for an object label.
void	SetObjectLabelOpacity (int index, byte opacity)
	Sets the label opacity for an object label.
byte	GetObjectLabelOpacity (int index)
	Gets the label opacity for an object label.
bool	HasObjectLabeling (int imageIndex)
	Whether the image is labeled for object detection and use with ELocator.
int	GetNumImagesWithObjectLabel (int labelIdx)
	Number of images that contain an object with the given label.
int	GetNumImagesWithObjectLabel (string label)
	Number of images that contain an object with the given label.
ulong	GetNumObjectsWithLabel (int labelIdx)
	Number of objects with the given label.
ulong	GetNumObjectsWithLabel (string label)
	Number of objects with the given label.
bool	HasOCRAnnotations (int imgId)
	Whether the specified image has OCR annotations.
void	UnsetOCRAnnotations (int imgId)
	Unsets the OCR labeling for the specified image.
eVision.EasyDeepOCR.ETextLine[]	GetOCRAnnotations (int imgId)
	Ground truth text lines for the specified image.
void	SetOCRAnnotations (int imgId, eVision.EasyDeepOCR.ETextLine[] texts)
	Sets the ground truth text lines for the specified image.
void	SplitDataset (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2, float proportion, bool random)
	Splits the dataset in two parts to be used for training and validation respectively.
void	SplitDataset (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2, float proportion)
	Splits the dataset in two parts to be used for training and validation respectively.
void	SplitDataset (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2)
	Splits the dataset in two parts to be used for training and validation respectively.
void	SplitDatasetForSegmentation (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2, float proportion, bool random)
	Splits the dataset in two parts for a supervised segmenter. The two parts are to be used for training and validation respectively.
void	SplitDatasetForSegmentation (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2, float proportion)
	Splits the dataset in two parts for a supervised segmenter. The two parts are to be used for training and validation respectively.
void	SplitDatasetForSegmentation (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2)
	Splits the dataset in two parts for a supervised segmenter. The two parts are to be used for training and validation respectively.
void	SplitDatasetForLocator (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2, float proportion, bool random)
	Splits the dataset in two parts for training and validation of a ELocator tool. Images without object labeling are excluded from the split.
void	SplitDatasetForLocator (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2, float proportion)
	Splits the dataset in two parts for training and validation of a ELocator tool. Images without object labeling are excluded from the split.
void	SplitDatasetForLocator (eVision.EasyDeepLearning.EDataset d1, eVision.EasyDeepLearning.EDataset d2)
	Splits the dataset in two parts for training and validation of a ELocator tool. Images without object labeling are excluded from the split.
eVision.EasyDeepLearning.EDatasetSplit	GetSplit (float trainingProportion, float validationPropotion, eVision.EasyDeepLearning.EDeepLearningToolType toolType, string goodLabel)
	Generates a split.
eVision.EasyDeepLearning.EDatasetSplit	GetSplit (float trainingProportion, float validationPropotion, eVision.EasyDeepLearning.EDeepLearningToolType toolType, string goodLabel, uint seed)
	Generates a split.
eVision.EasyDeepLearning.EDatasetSplit	GetSplit (int numTrainingImages, int numValidationImages, eVision.EasyDeepLearning.EDeepLearningToolType toolType, string goodLabel)
	Generates a split.
eVision.EasyDeepLearning.EDatasetSplit	GetSplit (int numTrainingImages, int numValidationImages, eVision.EasyDeepLearning.EDeepLearningToolType toolType, string goodLabel, uint seed)
	Generates a split.
eVision.EasyDeepLearning.EDataset	ExtractSplit (eVision.EasyDeepLearning.EDatasetSplit split, eVision.EasyDeepLearning.EDatasetType type)
	Creates a new EDataset containing only the images of the given split type.
void	SetRegionOfInterest (int imageIndex, int xOrg, int yOrg, int width, int height)
	Sets the region of interest for the specified image.
int	GetRegionOfInterestOriginX (int imageIndex)
	Region of interest origin abscissa for the specified image.
int	GetRegionOfInterestOriginY (int imageIndex)
	Region of interest origin ordinate for the specified image.
int	GetRegionOfInterestWidth (int imageIndex)
	Region of interest width for the specified image.
int	GetRegionOfInterestHeight (int imageIndex)
	Region of interest height for the specified image.
eVision.ERegion	GetMask (int imageIndex)
	Gets the mask region/don't care area for the given image. The mask is defined with respect to the region of interest of the image.
void	SetMask (int imageIndex, eVision.ERegion mask)
	Sets the mask region/don't care area for the given image. The mask is defined with respect to the region of interest of the image.

Properties
string	BasePath `[set]`
	Sets the base path for the images specified with a relative path.
uint	Width `[get]`
	Width of the region of interest of the first image added to the dataset.
uint	Height `[get]`
	Height of the region of interest of the first image added to the dataset.
uint	Channels `[get]`
	Number of channels of the first image added to the dataset.
int	NumImages `[get]`
	Number of images in the dataset.
int	NumImageFiles `[get]`
	Number of different image files contained in the dataset.
int	NumLabeledImages `[get]`
	Number of images that have a classification label.
int	NumUnlabeledImages `[get]`
	Number of images that doesn't have any classification label.
int[]	LabeledImagesIndexes `[get]`
	Gets a list of index corresponding to the images that have a classification label.
int[]	TextLabeledImagesIndexes `[get]`
	Gets a list of index corresponding to the images that have a text only label.
int[]	ImagesIndexesWithNoLabel `[get]`
	Gets a list of index corresponding to the images with no classification labels.
int	NumLabels `[get]`
	Number of labels in the dataset.
int	NumSegmentationLabels `[get]`
	Number of segmentation labels. A dataset has always at least one segmentation label: "background".
int	NumImagesWithSegmentation `[get]`
	Number of images that have a ground truth segmentation.
int	NumImagesWithoutSegmentation `[get]`
	Number of images that doesn't have a ground truth segmentation.
int	NumImagesWithoutForegroundSegments `[get]`
	Number of images that have a ground truth segmentation that has no foreground segments and is thus entirely composed of background pixels.
int	NumImagesWithForegroundSegments `[get]`
	Number of images that have a ground truth segmentation that has foreground segments and is this not entirely composed of background pixels.
int	NumObjectLabels `[get]`
	Number of object labels.
int	NumImagesWithObjectLabeling `[get]`
	Number of images in the dataset that are labeled for object detection.
int	NumImagesWithoutObjectLabeling `[get]`
	Number of images in the dataset that are not labeled for object detection.
int	NumImagesWithoutObjects `[get]`
	Number of images in the dataset that are labeled for object detection but have been associated with no object.
int	NumImagesWithObjects `[get]`
	Number of images in the dataset that are labeled for object detection and have 1 or more objects.
float	SameLabelMaxOverlap `[get]`
	Maximum overlap between objects with the same label in the dataset.
float	AbsoluteMaxOverlap `[get]`
	Absolute maximum overlap between objects in the dataset.
int	MaxNumObjectPerImage `[get]`
	Maximum number of objects for an image in the dataset. If no images has object labeling (see EDataset.HasObjectLabeling), the method returns -1.
int	ObjectSize `[get, set]`
	Object size for EasyLocate Interest point.
int	NumImagesWithOCR `[get]`
	Number of images with OCR annotations.
int	NumImagesWithoutOCR `[get]`
	Number of images without OCR annotations.
int	NumTexts `[get]`
	Number of annotated texts in the dataset.
eVision.EasyDeepLearning.EDataset	TextDataset `[get]`
	Transforms the current dataset to have only subwindow containing the actual text to read.
eVision.EasyDeepLearning.EDataset	TextOrientationDataset `[get]`
	Transforms the current text dataset to duplicate it with half image flipped, adding corresponding ground truth.

Detailed Description

EDataset manages a dataset of images.

A dataset is a collection of images with different types of labeling: labeling for the classification of images, labeling for the segmentation of pixels, and/or labeling for the detection of objects (classification and localization).

The dataset maintains 3 sets of labels for each type of labeling:

the classification labels that characterize an entire image;
the segmentation labels that characterize the pixels of an image; and
the object labels that characterize axis-aligned rectangle region of an image.

The classification and segmentation labels are entirely user defined. The set of segmentation labels will always contain at least the "Background" label representing pixels of the images that have no relevant information for your task (for example, in a defect segmentation application, the "Background" pixels would be the pixels without any defects).

For each type of labeling, an image can either be unlabeled or labeled. When an image in unlabeled for a given type of labeling, the image won't be used for training a deep learning tool that requires this type of labeling.

The image in the dataset can be stored as path to an image file or as an eVision image structure. Supported structures are 8-bits monochrome (EImageBW8), 16-bits monochrome (EImageBW16), and 24-bits color (EROIC24, EImageC24).

The dataset associates with each image a region of interest and a mask/don't care area. By default, the region of interest of an image is its full extent and its mask is empty.

A dataset can contain images with different sizes (with and height of their region of interest). However, the dataset has a default resolution (see EDataset.Width and EDataset.Height) that is used by deep learning tools that require the same input image size such as the EClassifier. When the images have different sizes, the default resolution will be the resolution of the region of interest of the first image added to the dataset.

License(s): BasicTypes

Constructor & Destructor Documentation

◆ EDataset() [1/2]

eVision.EasyDeepLearning.EDataset.EDataset ( )

Constructs a EDataset object.

◆ EDataset() [2/2]

eVision.EasyDeepLearning.EDataset.EDataset ( eVision.EasyDeepLearning.EDataset other )

Copy constructor of a EDataset object.

Parameters

other Reference to the EDataset object that should be copied

Member Function Documentation

◆ AddImage() [1/8]

int eVision.EasyDeepLearning.EDataset.AddImage ( eVision.EBaseROI img )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

img The image

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImage() [2/8]

int eVision.EasyDeepLearning.EDataset.AddImage	(	eVision.EBaseROI	img,
		int	originX,
		int	originY,
		int	width,
		int	height,
		eVision.ERegion	mask )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

img	The image
originX	Region of interest origin abscissa
originY	Region of interest origin ordinate
width	Region of interest width
height	Region of interest height
mask	The mask for the image (with respect to the region of interest)

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImage() [3/8]

int eVision.EasyDeepLearning.EDataset.AddImage	(	eVision.EBaseROI	img,
		int	originX,
		int	originY,
		int	width,
		int	height,
		eVision.ERegion	mask,
		string	label )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

img	The image
originX	Region of interest origin abscissa
originY	Region of interest origin ordinate
width	Region of interest width
height	Region of interest height
mask	The mask for the image (with respect to the region of interest)
label	The label

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImage() [4/8]

int eVision.EasyDeepLearning.EDataset.AddImage	(	eVision.EBaseROI	img,
		string	label )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

img	The image
label	The label

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImage() [5/8]

int eVision.EasyDeepLearning.EDataset.AddImage ( string imagePath )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

imagePath The path to an image

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImage() [6/8]

int eVision.EasyDeepLearning.EDataset.AddImage	(	string	imagePath,
		int	originX,
		int	originY,
		int	width,
		int	height,
		eVision.ERegion	mask )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

imagePath	The path to an image
originX	Region of interest origin abscissa
originY	Region of interest origin ordinate
width	Region of interest width
height	Region of interest height
mask	The mask for the image (with respect to the region of interest)

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImage() [7/8]

int eVision.EasyDeepLearning.EDataset.AddImage	(	string	imagePath,
		int	originX,
		int	originY,
		int	width,
		int	height,
		eVision.ERegion	mask,
		string	label )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

imagePath	The path to an image
originX	Region of interest origin abscissa
originY	Region of interest origin ordinate
width	Region of interest width
height	Region of interest height
mask	The mask for the image (with respect to the region of interest)
label	The label

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImage() [8/8]

int eVision.EasyDeepLearning.EDataset.AddImage	(	string	imagePath,
		string	label )

Adds an image to the dataset.

The image can be specified by its path on the filesystem (parameter imagePath) or by an eVision image buffer (parameter img).

By default, an image will have no classification label and no segmentation. The label of the image can be directly specified when adding the image to the dataset. If the given label was not in the classification labels of the dataset, it will be automatically added to them.

If no region of interest and/or mask is specified, the region of interest of the image will be its full extent and its mask will be empty.

The method returns the index of the image in the dataset. This index is equal to EDataset.NumImages - 1 and can be used to add, remove, or modify the annotations for the image.

Parameters

imagePath	The path to an image
label	The label

When adding eVision images (EBaseROI) to a dataset, the dataset will retain a copy of the image.

When specifying an individual region of interest and/or mask, the dataset will retain a copy of these.

◆ AddImageObject() [1/2]

void eVision.EasyDeepLearning.EDataset.AddImageObject	(	int	imageIndex,
		eVision.EasyDeepLearning.ELocatorObject	obj )

Adds an object to the image.

Parameters

imageIndex	Index of the image
obj	Object

The image will be marked as labeled for object detection (EDataset.HasObjectLabeling equals to true) after a call to this method.

The object must be located at least partially within the region of interest of the image. Otherwise, when the object is completely outside of the region of interest, an exception will be thrown.

If the label of the object does not exist in the object labels of the dataset, the label will be added to the object labels of the dataset.

◆ AddImageObject() [2/2]

void eVision.EasyDeepLearning.EDataset.AddImageObject	(	int	imageIndex,
		string	label,
		eVision.ERectangleRegion	box )

Adds an object to the image.

Parameters

imageIndex	Index of the image
label	Label of the object
box	Axis-aligned rectangle

The image will be marked as labeled for object detection (EDataset.HasObjectLabeling equals to true) after a call to this method.

The object must be located at least partially within the region of interest of the image. Otherwise, when the object is completely outside of the region of interest, an exception will be thrown.

If the label of the object does not exist in the object labels of the dataset, the label will be added to the object labels of the dataset.

◆ AddImages() [1/4]

int eVision.EasyDeepLearning.EDataset.AddImages ( string filter )

Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.

By default, all the images will have no classification label and no classification. However, a label can be directly specified for all of the images.

The method returns the number of images added to the dataset.

Parameters

filter A glob filter

The filter is a glob pattern. This means the wildcard characters "*" and "?" correspond to "zero or more character" and "a single character" respectively. For example, the filter "*_good_*.png" will match any filename that contains the string "_good_" and has a png extension.

◆ AddImages() [2/4]

int eVision.EasyDeepLearning.EDataset.AddImages	(	string	filter,
		int	originX,
		int	originY,
		int	width,
		int	height,
		eVision.ERegion	mask )

Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.

By default, all the images will have no classification label and no classification. However, a label can be directly specified for all of the images.

The method returns the number of images added to the dataset.

Parameters

filter	A glob filter
originX	Region of interest origin abscissa
originY	Region of interest origin ordinate
width	Region of interest width
height	Region of interest height
mask	The mask for the images

The filter is a glob pattern. This means the wildcard characters "*" and "?" correspond to "zero or more character" and "a single character" respectively. For example, the filter "*_good_*.png" will match any filename that contains the string "_good_" and has a png extension.

◆ AddImages() [3/4]

int eVision.EasyDeepLearning.EDataset.AddImages	(	string	filter,
		int	originX,
		int	originY,
		int	width,
		int	height,
		eVision.ERegion	mask,
		string	label )

Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.

By default, all the images will have no classification label and no classification. However, a label can be directly specified for all of the images.

The method returns the number of images added to the dataset.

Parameters

filter	A glob filter
originX	Region of interest origin abscissa
originY	Region of interest origin ordinate
width	Region of interest width
height	Region of interest height
mask	The mask for the images
label	A label.

The filter is a glob pattern. This means the wildcard characters "*" and "?" correspond to "zero or more character" and "a single character" respectively. For example, the filter "*_good_*.png" will match any filename that contains the string "_good_" and has a png extension.

◆ AddImages() [4/4]

int eVision.EasyDeepLearning.EDataset.AddImages	(	string	filter,
		string	label )

Adds all the images present in the directory specified by the parameter path and whose filename matches the filter.

By default, all the images will have no classification label and no classification. However, a label can be directly specified for all of the images.

The method returns the number of images added to the dataset.

Parameters

filter	A glob filter
label	A label.

The filter is a glob pattern. This means the wildcard characters "*" and "?" correspond to "zero or more character" and "a single character" respectively. For example, the filter "*_good_*.png" will match any filename that contains the string "_good_" and has a png extension.

◆ AddImagesFromCSV() [1/5]

int eVision.EasyDeepLearning.EDataset.AddImagesFromCSV ( string csvFile )

Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.

If the image path in the CSV file are relative, we use the directory containing the given CSV file as the root of the image paths.

The method returns the number of images added to the dataset.

◆ AddImagesFromCSV() [2/5]

int eVision.EasyDeepLearning.EDataset.AddImagesFromCSV	(	string	csvFile,
		int	imageColumn )

Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.

If the image path in the CSV file are relative, we use the directory containing the given CSV file as the root of the image paths.

The method returns the number of images added to the dataset.

◆ AddImagesFromCSV() [3/5]

int eVision.EasyDeepLearning.EDataset.AddImagesFromCSV	(	string	csvFile,
		int	imageColumn,
		int	labelColumn )

Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.

If the image path in the CSV file are relative, we use the directory containing the given CSV file as the root of the image paths.

The method returns the number of images added to the dataset.

◆ AddImagesFromCSV() [4/5]

int eVision.EasyDeepLearning.EDataset.AddImagesFromCSV	(	string	csvFile,
		int	imageColumn,
		int	labelColumn,
		int	skipLines )

Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.

If the image path in the CSV file are relative, we use the directory containing the given CSV file as the root of the image paths.

The method returns the number of images added to the dataset.

◆ AddImagesFromCSV() [5/5]

int eVision.EasyDeepLearning.EDataset.AddImagesFromCSV	(	string	csvFile,
		int	imageColumn,
		int	labelColumn,
		int	skipLines,
		string	separator )

Adds images using a CSV (Comma Separated Value) file where each row correspond to an image and the file path and label of each image is given in the columns imageColumn and labelColumn respectively.

If the image path in the CSV file are relative, we use the directory containing the given CSV file as the root of the image paths.

The method returns the number of images added to the dataset.

◆ AddLabel()

void eVision.EasyDeepLearning.EDataset.AddLabel ( string label )

Adds a label to the dataset.

Parameters

label Name of the new label

◆ AddObjectLabel() [1/2]

void eVision.EasyDeepLearning.EDataset.AddObjectLabel ( string label )

Adds an object label.

Parameters

label Label to add

◆ AddObjectLabel() [2/2]

void eVision.EasyDeepLearning.EDataset.AddObjectLabel	(	string	label,
		float	weight )

Adds an object label.

Parameters

label	Label to add
weight	Weight of the label

◆ AddRegionToSegment() [1/2]

void eVision.EasyDeepLearning.EDataset.AddRegionToSegment	(	int	imageIndex,
		int	segmentationLabelIndex,
		eVision.ERegion	region )

Adds a region (see ERegion) to the given segment of an image. If, before this call, the image had no segmentation, the pixels not in the given region will be set to the background segmentation label.

Parameters

imageIndex	Image index
segmentationLabelIndex	Segmentation label index
region	Region to add

◆ AddRegionToSegment() [2/2]

void eVision.EasyDeepLearning.EDataset.AddRegionToSegment	(	int	imageIndex,
		string	label,
		eVision.ERegion	region )

Adds a region (see ERegion) to the given segment of an image. If, before this call, the image had no segmentation, the pixels not in the given region will be set to the background segmentation label.

Parameters

imageIndex	Image index
label	Segmentation label
region	Region to add

◆ AddSegmentationLabel() [1/2]

int eVision.EasyDeepLearning.EDataset.AddSegmentationLabel ( string label )

Adds a segmentation label. The index of the new segmentation labels is returned by the method.

Parameters

label Name of the segmentation label

◆ AddSegmentationLabel() [2/2]

int eVision.EasyDeepLearning.EDataset.AddSegmentationLabel	(	string	label,
		float	labelWeight )

Adds a segmentation label. The index of the new segmentation labels is returned by the method.

Parameters

label	Name of the segmentation label
labelWeight	Weight of the segmentation label

◆ Clear()

virtual void eVision.EasyDeepLearning.EDataset.Clear ( )

virtual

Clears the dataset.

Reimplemented in eVision.EasyDeepLearning.EClassificationDataset.

◆ Export() [1/12]

void eVision.EasyDeepLearning.EDataset.Export ( string directory )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory A string containing the full path to the directory.

◆ Export() [2/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		bool	keepFilename )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
keepFilename	Keep the original filename of the images

◆ Export() [3/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		bool	keepFilename,
		eVision.EImageFileType	fileType )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
keepFilename	Keep the original filename of the images
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.

◆ Export() [4/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		bool	keepFilename,
		eVision.EImageFileType	fileType,
		int	quality )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
keepFilename	Keep the original filename of the images
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.
quality	Quality or compression parameters for EBaseROI.SavePng, EBaseROI.SaveJpeg, or EBaseROI.SaveJpeg2K. A value of -1 means the default value.

◆ Export() [5/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		eVision.EImageFileType	fileType )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.

◆ Export() [6/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		eVision.EImageFileType	fileType,
		int	quality )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.
quality	Quality or compression parameters for EBaseROI.SavePng, EBaseROI.SaveJpeg, or EBaseROI.SaveJpeg2K. A value of -1 means the default value.

◆ Export() [7/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		int	width,
		int	height,
		int	channels )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
width	Width of the image.
height	Height of the image.
channels	Number of channels of the image (only 1 and 3 are valid, for grayscale and RGB respectively).

◆ Export() [8/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		int	width,
		int	height,
		int	channels,
		bool	keepFilename )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
width	Width of the image.
height	Height of the image.
channels	Number of channels of the image (only 1 and 3 are valid, for grayscale and RGB respectively).
keepFilename	Keep the original filename of the images

◆ Export() [9/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		int	width,
		int	height,
		int	channels,
		bool	keepFilename,
		eVision.EImageFileType	fileType )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
width	Width of the image.
height	Height of the image.
channels	Number of channels of the image (only 1 and 3 are valid, for grayscale and RGB respectively).
keepFilename	Keep the original filename of the images
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.

◆ Export() [10/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		int	width,
		int	height,
		int	channels,
		bool	keepFilename,
		eVision.EImageFileType	fileType,
		int	quality )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
width	Width of the image.
height	Height of the image.
channels	Number of channels of the image (only 1 and 3 are valid, for grayscale and RGB respectively).
keepFilename	Keep the original filename of the images
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.
quality	Quality or compression parameters for EBaseROI.SavePng, EBaseROI.SaveJpeg, or EBaseROI.SaveJpeg2K. A value of -1 means the default value.

◆ Export() [11/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		int	width,
		int	height,
		int	channels,
		eVision.EImageFileType	fileType )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
width	Width of the image.
height	Height of the image.
channels	Number of channels of the image (only 1 and 3 are valid, for grayscale and RGB respectively).
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.

◆ Export() [12/12]

void eVision.EasyDeepLearning.EDataset.Export	(	string	directory,
		int	width,
		int	height,
		int	channels,
		eVision.EImageFileType	fileType,
		int	quality )

Exports the dataset and its images to the given directory.

A new EDataset object with relative paths to the images is saved into the given directory.
The exported images will be placed in a sub directory named "Images".

By default, the images will be saved under the filename "Image_[id].[ext]" where [id] is the index of the image in the dataset and [ext] is the image extension. If fileType is set to EImageFileType_Auto, [ext] will be the same as the original image. Otherwise, [ext] will be deduced from the specified file type. If keepFilename is set to true, the images will keep their original filename (except for their extensions). In two images have the same filename, an index will be automatically added to avoid nay conflict. A width, height, and number of channels can be optionally specified to reformat all the images in the dataset. Note that, in this case, only the ROI of the image will be exported.

Parameters

directory	A string containing the full path to the directory.
width	Width of the image.
height	Height of the image.
channels	Number of channels of the image (only 1 and 3 are valid, for grayscale and RGB respectively).
fileType	File type for the exported file. If EImageFileType_Auto, the same file type as the original image is used.
quality	Quality or compression parameters for EBaseROI.SavePng, EBaseROI.SaveJpeg, or EBaseROI.SaveJpeg2K. A value of -1 means the default value.

◆ ExtractSplit()

eVision.EasyDeepLearning.EDataset eVision.EasyDeepLearning.EDataset.ExtractSplit	(	eVision.EasyDeepLearning.EDatasetSplit	split,
		eVision.EasyDeepLearning.EDatasetType	type )

Creates a new EDataset containing only the images of the given split type.

Parameters

split	Split
type	Type of the split to extract

◆ GetAvailableImageAnnotationFormat() [1/2]

eVision.EasyDeepLearning.EImageAnnotationFormat eVision.EasyDeepLearning.EDataset.GetAvailableImageAnnotationFormat ( int imgId )

Image annotation file formats that are available next to the image file.

Parameters

imgId Index of an image

◆ GetAvailableImageAnnotationFormat() [2/2]

eVision.EasyDeepLearning.EImageAnnotationFormat eVision.EasyDeepLearning.EDataset.GetAvailableImageAnnotationFormat ( string imageFile )

Image annotation file formats that are available next to the image file.

Parameters

imageFile Path to an image file

◆ GetImageCopy() [1/2]

eVision.EBaseROI eVision.EasyDeepLearning.EDataset.GetImageCopy ( int index )

Gets a copy of the i-th image of the dataset.

Parameters

index The index of the image.

◆ GetImageCopy() [2/2]

void eVision.EasyDeepLearning.EDataset.GetImageCopy	(	int	index,
		eVision.EBaseROI	img )

Gets a copy of the i-th image of the dataset.

Parameters

index	The index of the image.
img	Image object to copy the image into.

◆ GetImageLabel()

string eVision.EasyDeepLearning.EDataset.GetImageLabel ( int index )

Gets the label of the i-th image of the dataset. An exception is thrown if the image has no classification label.

Parameters

index The index of the image for which to get the label.

◆ GetImageNumObjects()

int eVision.EasyDeepLearning.EDataset.GetImageNumObjects ( int imageIndex )

Number of objects for the specified image.

Parameters

imageIndex Index of the image

◆ GetImageObject()

eVision.EasyDeepLearning.ELocatorObject eVision.EasyDeepLearning.EDataset.GetImageObject	(	int	imageIndex,
		int	objectIndex )

Gets the specified object for the given image.

Parameters

imageIndex	Index of the image
objectIndex	Index of the object between 0 and EDataset.GetImageNumObjects

◆ GetImageObjects()

eVision.EasyDeepLearning.ELocatorObject[] eVision.EasyDeepLearning.EDataset.GetImageObjects ( int imageIndex )

Gets all the objects of the specified image.

Parameters

imageIndex Index of the image

◆ GetImagePath()

string eVision.EasyDeepLearning.EDataset.GetImagePath ( int index )

Gets the path of the i-th image of the dataset. If the image was not given using a path, the method will throw an exception.

Parameters

index The index of the image for which to get the path.

◆ GetImagesIndexesWithLabel() [1/2]

uint[] eVision.EasyDeepLearning.EDataset.GetImagesIndexesWithLabel ( int labelIndex )

Gets a list of index corresponding to the images associated with the given label.

Parameters

labelIndex The index of the label (starting from 0 to EDataset.NumLabels - 1)

◆ GetImagesIndexesWithLabel() [2/2]

uint[] eVision.EasyDeepLearning.EDataset.GetImagesIndexesWithLabel ( string label )

Gets a list of index corresponding to the images associated with the given label.

Parameters

label The label

◆ GetLabel()

string eVision.EasyDeepLearning.EDataset.GetLabel ( int i )

Gets the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1).

Parameters

i	Label index

◆ GetLabelColor()

eVision.ERGBColor eVision.EasyDeepLearning.EDataset.GetLabelColor ( int index )

Gets the label color for a classification label.

Parameters

index Label index (0 to EDataset.NumLabels - 1).

◆ GetLabelOpacity()

byte eVision.EasyDeepLearning.EDataset.GetLabelOpacity ( int index )

Gets the label opacity for a classification label.

Parameters

index Label index (0 to EDataset.NumLabels - 1).

◆ GetLabelWeight()

float eVision.EasyDeepLearning.EDataset.GetLabelWeight ( int index )

Gets the weight associated to the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1).

Parameters

index Label index

◆ GetMask()

eVision.ERegion eVision.EasyDeepLearning.EDataset.GetMask ( int imageIndex )

Gets the mask region/don't care area for the given image. The mask is defined with respect to the region of interest of the image.

Parameters

imageIndex Index of the image for which to get the mask region

◆ GetNumImagesForFile()

int eVision.EasyDeepLearning.EDataset.GetNumImagesForFile ( string filePath )

Get the number of images in the dataset that are associated with the given image file.

Parameters

filePath the path to the image.

The number of image files can be different than the number of images of the dataset. An image file can correspond to several dataset images (with different ROI).

◆ GetNumImagesForSegmentationLabel() [1/2]

int eVision.EasyDeepLearning.EDataset.GetNumImagesForSegmentationLabel ( int labelId )

Number of images containing the given segmentation label.

Parameters

labelId Index of the segmentation label

◆ GetNumImagesForSegmentationLabel() [2/2]

int eVision.EasyDeepLearning.EDataset.GetNumImagesForSegmentationLabel ( string label )

Number of images containing the given segmentation label.

Parameters

label Segmentation label

◆ GetNumImagesWithObjectLabel() [1/2]

int eVision.EasyDeepLearning.EDataset.GetNumImagesWithObjectLabel ( int labelIdx )

Number of images that contain an object with the given label.

Parameters

labelIdx Index of the object label.

◆ GetNumImagesWithObjectLabel() [2/2]

int eVision.EasyDeepLearning.EDataset.GetNumImagesWithObjectLabel ( string label )

Number of images that contain an object with the given label.

Parameters

label Label.

◆ GetNumObjectsWithLabel() [1/2]

ulong eVision.EasyDeepLearning.EDataset.GetNumObjectsWithLabel ( int labelIdx )

Number of objects with the given label.

Parameters

labelIdx Index of the object label.

◆ GetNumObjectsWithLabel() [2/2]

ulong eVision.EasyDeepLearning.EDataset.GetNumObjectsWithLabel ( string label )

Number of objects with the given label.

Parameters

label Label.

◆ GetNumPixelsForSegmentationLabel() [1/2]

ulong eVision.EasyDeepLearning.EDataset.GetNumPixelsForSegmentationLabel ( int labelId )

Number of pixels assigned to the given segmentation label.

Parameters

labelId Index of the segmentation label

◆ GetNumPixelsForSegmentationLabel() [2/2]

ulong eVision.EasyDeepLearning.EDataset.GetNumPixelsForSegmentationLabel ( string label )

Number of pixels assigned to the given segmentation label.

Parameters

label Segmentation label

◆ GetNumSegmentedBlobs() [1/3]

int eVision.EasyDeepLearning.EDataset.GetNumSegmentedBlobs ( int imageId )

Number of segmented blobs in the image for all non-background labels or for a specific label.

Parameters

imageId Image index

A segmented blob is a contiguous area assigned to a label different than "Background".

◆ GetNumSegmentedBlobs() [2/3]

int eVision.EasyDeepLearning.EDataset.GetNumSegmentedBlobs	(	int	imageId,
		int	labelId )

Number of segmented blobs in the image for all non-background labels or for a specific label.

Parameters

imageId	Image index
labelId	Label index

A segmented blob is a contiguous area assigned to a label different than "Background".

◆ GetNumSegmentedBlobs() [3/3]

int eVision.EasyDeepLearning.EDataset.GetNumSegmentedBlobs	(	int	imageId,
		string	label )

Number of segmented blobs in the image for all non-background labels or for a specific label.

Parameters

imageId	Image index
label	Label

A segmented blob is a contiguous area assigned to a label different than "Background".

◆ GetObjectLabel()

string eVision.EasyDeepLearning.EDataset.GetObjectLabel ( int labelIndex )

Object label.

Parameters

labelIndex Index of the object label between 0 and EDataset.NumObjectLabels

◆ GetObjectLabelColor()

eVision.ERGBColor eVision.EasyDeepLearning.EDataset.GetObjectLabelColor ( int index )

Gets the label color for an object label.

Parameters

index Label index (0 to EDataset.NumLabels - 1).

◆ GetObjectLabelOpacity()

byte eVision.EasyDeepLearning.EDataset.GetObjectLabelOpacity ( int index )

Gets the label opacity for an object label.

Parameters

index Label index (0 to EDataset.NumLabels - 1).

◆ GetObjectLabelWeight() [1/2]

float eVision.EasyDeepLearning.EDataset.GetObjectLabelWeight ( int labelIndex )

Gets the weight of an object label.

Parameters

labelIndex Index of the object label

◆ GetObjectLabelWeight() [2/2]

float eVision.EasyDeepLearning.EDataset.GetObjectLabelWeight ( string label )

Gets the weight of an object label.

Parameters

label Label

◆ GetOCRAnnotations()

eVision.EasyDeepOCR.ETextLine[] eVision.EasyDeepLearning.EDataset.GetOCRAnnotations ( int imgId )

Ground truth text lines for the specified image.

Parameters

imgId Index of the image.

◆ GetRegionForSegment() [1/2]

eVision.ERegion eVision.EasyDeepLearning.EDataset.GetRegionForSegment	(	int	imageIndex,
		int	segmentationLabelIndex )

Gets a region corresponding to the pixels of the given segment.

Parameters

imageIndex	Image index
segmentationLabelIndex	Segmentation label index

◆ GetRegionForSegment() [2/2]

eVision.ERegion eVision.EasyDeepLearning.EDataset.GetRegionForSegment	(	int	imageIndex,
		string	segmentationLabel )

Gets a region corresponding to the pixels of the given segment.

Parameters

imageIndex	Image index
segmentationLabel	Segmentation label

◆ GetRegionOfInterestHeight()

int eVision.EasyDeepLearning.EDataset.GetRegionOfInterestHeight ( int imageIndex )

Region of interest height for the specified image.

Parameters

imageIndex Index of the image.

◆ GetRegionOfInterestOriginX()

int eVision.EasyDeepLearning.EDataset.GetRegionOfInterestOriginX ( int imageIndex )

Region of interest origin abscissa for the specified image.

Parameters

imageIndex Index of the image.

◆ GetRegionOfInterestOriginY()

int eVision.EasyDeepLearning.EDataset.GetRegionOfInterestOriginY ( int imageIndex )

Region of interest origin ordinate for the specified image.

Parameters

imageIndex Index of the image.

◆ GetRegionOfInterestWidth()

int eVision.EasyDeepLearning.EDataset.GetRegionOfInterestWidth ( int imageIndex )

Region of interest width for the specified image.

Parameters

imageIndex Index of the image.

◆ GetSegmentationLabel()

string eVision.EasyDeepLearning.EDataset.GetSegmentationLabel ( int index )

Gets the segmentation label.

Parameters

index Index of the segmentation label

The segmentation label index 0 is reserved and corresponds to the "background" label. This segmentation label can't be changed.

◆ GetSegmentationLabelColor()

eVision.ERGBColor eVision.EasyDeepLearning.EDataset.GetSegmentationLabelColor ( int index )

Gets the label color for a segmentation label.

Parameters

index Label index (0 to EDataset.NumLabels - 1).

◆ GetSegmentationLabelOpacity()

byte eVision.EasyDeepLearning.EDataset.GetSegmentationLabelOpacity ( int index )

Gets the label opacity for a segmentation label.

Parameters

index Label index (0 to EDataset.NumLabels - 1).

◆ GetSegmentationLabelWeight() [1/2]

float eVision.EasyDeepLearning.EDataset.GetSegmentationLabelWeight ( int index )

Gets the segmentation label weights.

Parameters

index Index of the segmentation label

◆ GetSegmentationLabelWeight() [2/2]

float eVision.EasyDeepLearning.EDataset.GetSegmentationLabelWeight ( string label )

Gets the segmentation label weights.

Parameters

label Segmentation label

◆ GetSegmentationMap()

eVision.EImageBW16 eVision.EasyDeepLearning.EDataset.GetSegmentationMap ( int imageIndex )

Gets the segmentation map of an image. The segmentation map is a 16-bit EROIBW16 image where the value of each pixel is equal to the corresponding segmentation label index (see EDataset.GetSegmentationLabel).

If an image has no segmentation (EDataset.HasSegmentation), the getter will throw an exception.

Parameters

imageIndex Image index

◆ GetSplit() [1/4]

eVision.EasyDeepLearning.EDatasetSplit eVision.EasyDeepLearning.EDataset.GetSplit	(	float	trainingProportion,
		float	validationPropotion,
		eVision.EasyDeepLearning.EDeepLearningToolType	toolType,
		string	goodLabel )

Generates a split.

Parameters

trainingProportion	Approximate proportion of training images
validationPropotion	Approximate proportion of validation images
toolType	Tool type for which to generate the split
goodLabel	The good label for EasySegment Unsupervised split

◆ GetSplit() [2/4]

eVision.EasyDeepLearning.EDatasetSplit eVision.EasyDeepLearning.EDataset.GetSplit	(	float	trainingProportion,
		float	validationPropotion,
		eVision.EasyDeepLearning.EDeepLearningToolType	toolType,
		string	goodLabel,
		uint	seed )

Generates a split.

Parameters

trainingProportion	Approximate proportion of training images
validationPropotion	Approximate proportion of validation images
toolType	Tool type for which to generate the split
goodLabel	The good label for EasySegment Unsupervised split
seed	Seed for randomization

◆ GetSplit() [3/4]

eVision.EasyDeepLearning.EDatasetSplit eVision.EasyDeepLearning.EDataset.GetSplit	(	int	numTrainingImages,
		int	numValidationImages,
		eVision.EasyDeepLearning.EDeepLearningToolType	toolType,
		string	goodLabel )

Generates a split.

Parameters

numTrainingImages	Number of training images
numValidationImages	Number of validation images
toolType	Tool type for which to generate the split
goodLabel	The good label for EasySegment Unsupervised split

◆ GetSplit() [4/4]

eVision.EasyDeepLearning.EDatasetSplit eVision.EasyDeepLearning.EDataset.GetSplit	(	int	numTrainingImages,
		int	numValidationImages,
		eVision.EasyDeepLearning.EDeepLearningToolType	toolType,
		string	goodLabel,
		uint	seed )

Generates a split.

Parameters

numTrainingImages	Number of training images
numValidationImages	Number of validation images
toolType	Tool type for which to generate the split
goodLabel	The good label for EasySegment Unsupervised split
seed	Seed for randomization

◆ HasForegroundSegments()

bool eVision.EasyDeepLearning.EDataset.HasForegroundSegments ( int imageId )

Whether the image segmentation contains segments that are not background. If the image has no segmentation, this method throws an exception.

Parameters

imageId Image index

◆ HasLabel()

bool eVision.EasyDeepLearning.EDataset.HasLabel ( int index )

Whether an image has a classification label.

Parameters

index Index of an image.

◆ HasObjectLabeling()

bool eVision.EasyDeepLearning.EDataset.HasObjectLabeling ( int imageIndex )

Whether the image is labeled for object detection and use with ELocator.

Parameters

imageIndex Index of the image

◆ HasOCRAnnotations()

bool eVision.EasyDeepLearning.EDataset.HasOCRAnnotations ( int imgId )

Whether the specified image has OCR annotations.

Parameters

imgId Index of the image.

◆ HasSegmentation()

bool eVision.EasyDeepLearning.EDataset.HasSegmentation ( int imageId )

Whether the image has a segmentation.

Parameters

imageId Image index

◆ ImportPascalVOCXMLAnnotations() [1/2]

void eVision.EasyDeepLearning.EDataset.ImportPascalVOCXMLAnnotations ( )

Imports Pascal VOC XML annotations (for EasyLocate Bounding Box).

This method may add new labels to the object labels.

The version that do not take an image index attempts to import the annotation from all images currently in the dataset.

◆ ImportPascalVOCXMLAnnotations() [2/2]

void eVision.EasyDeepLearning.EDataset.ImportPascalVOCXMLAnnotations ( int imgId )

Imports Pascal VOC XML annotations (for EasyLocate Bounding Box).

This method may add new labels to the object labels.

The version that do not take an image index attempts to import the annotation from all images currently in the dataset.

Parameters

imgId The image index for which to import the annotation

◆ ImportYOLOTXTAnnotations() [1/2]

void eVision.EasyDeepLearning.EDataset.ImportYOLOTXTAnnotations	(	int	imgId,
		string	labelFile )

Imports YOLO TXT annotations (for EasyLocate Bounding Box).

You need to specify the list of labels associated with the YOLO TXT annotations through an array of string or through a file containing this list. This method may add new labels to the object labels.

The version that do not take an image index attempts to import the annotation from all images currently in the dataset.

Parameters

imgId	The image index for which to import the annotation
labelFile	Path to a file containing the label list

◆ ImportYOLOTXTAnnotations() [2/2]

void eVision.EasyDeepLearning.EDataset.ImportYOLOTXTAnnotations ( string labelFile )

Imports YOLO TXT annotations (for EasyLocate Bounding Box).

You need to specify the list of labels associated with the YOLO TXT annotations through an array of string or through a file containing this list. This method may add new labels to the object labels.

The version that do not take an image index attempts to import the annotation from all images currently in the dataset.

Parameters

labelFile Path to a file containing the label list

◆ IsEmbeddedImage()

bool eVision.EasyDeepLearning.EDataset.IsEmbeddedImage ( int index )

Whether the image is embedded into the dataset and is not a reference towards an image file.

Parameters

index Index of the image.

◆ IsImageFile()

bool eVision.EasyDeepLearning.EDataset.IsImageFile ( int index )

Whether the image is stored as a file path towards an image.

Parameters

index Index of the image.

◆ Load() [1/2]

void eVision.EasyDeepLearning.EDataset.Load ( eVision.ESerializer serializer )

Loads a classification dataset from disk.

Parameters

serializer The serializer.

◆ Load() [2/2]

void eVision.EasyDeepLearning.EDataset.Load ( string path )

Loads a classification dataset from disk.

Parameters

path	A string containing the full path to the dataset file.

◆ RemoveImage()

void eVision.EasyDeepLearning.EDataset.RemoveImage ( int imgId )

Removes the image at the given index. All annotations (label, segmentation, objects) will be lost.

Parameters

imgId Index of the image to remove

◆ RemoveImageObject()

void eVision.EasyDeepLearning.EDataset.RemoveImageObject	(	int	imageIndex,
		int	objectIndex )

Removes an object from an image.

Parameters

imageIndex	Index of the image
objectIndex	Index of the object between 0 and EDataset.GetImageNumObjects

◆ RemoveLabel() [1/2]

void eVision.EasyDeepLearning.EDataset.RemoveLabel ( int labelId )

Removes a label from the dataset.

All the images associated with this label will be set to unlabeled (see EDataset.HasLabel).

Parameters

labelId Index of the label to remove

◆ RemoveLabel() [2/2]

void eVision.EasyDeepLearning.EDataset.RemoveLabel ( string label )

Removes a label from the dataset.

All the images associated with this label will be set to unlabeled (see EDataset.HasLabel).

Parameters

label Name of the label to remove

◆ RemoveObjectLabel() [1/2]

void eVision.EasyDeepLearning.EDataset.RemoveObjectLabel ( int labelIndex )

Removes an object label.

Parameters

labelIndex Index of the object label to remove

◆ RemoveObjectLabel() [2/2]

void eVision.EasyDeepLearning.EDataset.RemoveObjectLabel ( string label )

Removes an object label.

Parameters

label Label to remove

◆ RemoveSegmentationLabel() [1/2]

void eVision.EasyDeepLearning.EDataset.RemoveSegmentationLabel ( int index )

Remove a segmentation label.

This method sets all the pixels in the dataset assigned to this label to the background label.

Parameters

index Index of the segmentation label to remove

◆ RemoveSegmentationLabel() [2/2]

void eVision.EasyDeepLearning.EDataset.RemoveSegmentationLabel ( string label )

Remove a segmentation label.

This method sets all the pixels in the dataset assigned to this label to the background label.

Parameters

label Name of the segmentation label to remove

◆ ResetImageObjectLabeling()

void eVision.EasyDeepLearning.EDataset.ResetImageObjectLabeling ( int imageIndex )

Resets the object labeling for the specified image. This sets the image as having been labeled for object detection with no object present in the image.

Parameters

imageIndex Index of the image

◆ ResetSegmentation()

void eVision.EasyDeepLearning.EDataset.ResetSegmentation ( int imageIndex )

Resets the segmentation of an image by setting all pixels to background.

Parameters

imageIndex Image index

◆ Save() [1/2]

void eVision.EasyDeepLearning.EDataset.Save ( eVision.ESerializer serializer )

Saves a classification dataset to disk, containing the file paths to the images in the dataset and their associated label.

Parameters

serializer The serializer.

This method only save the image that were given to this EDataset instance as EBaseROI pointers.

To obtain a portable EDataset file, please use EDataset.Export.

◆ Save() [2/2]

void eVision.EasyDeepLearning.EDataset.Save ( string path )

Saves a classification dataset to disk, containing the file paths to the images in the dataset and their associated label.

Parameters

path	A string containing the full path to the dataset file.

This method only save the image that were given to this EDataset instance as EBaseROI pointers.

To obtain a portable EDataset file, please use EDataset.Export.

◆ SaveWithRelativePath()

void eVision.EasyDeepLearning.EDataset.SaveWithRelativePath ( string path )

Saves the dataset with image paths relative to the directory in which the dataset file is saved.

◆ SetImageLabel() [1/2]

void eVision.EasyDeepLearning.EDataset.SetImageLabel	(	int	index,
		string	label )

Sets the label of images in the dataset.

Parameters

index	The index of the image for which to set the label.
label	The label

The filter is a glob pattern. This means the wildcard characters "*" and "?" correspond to "zero or more character" and "a single character" respectively. For example, the filter "*_good_*.png" will match any filename that contains the string "_good_" and has a png extension.

◆ SetImageLabel() [2/2]

void eVision.EasyDeepLearning.EDataset.SetImageLabel	(	string	filter,
		string	label )

Sets the label of images in the dataset.

Parameters

filter	A glob filter
label	The label

The filter is a glob pattern. This means the wildcard characters "*" and "?" correspond to "zero or more character" and "a single character" respectively. For example, the filter "*_good_*.png" will match any filename that contains the string "_good_" and has a png extension.

◆ SetImageObject() [1/3]

void eVision.EasyDeepLearning.EDataset.SetImageObject	(	int	imageIndex,
		int	objectIndex,
		eVision.EasyDeepLearning.ELocatorObject	obj )

Sets the specified object for the given image.

Parameters

imageIndex	Index of the image
objectIndex	Index of the object between 0 and EDataset.GetImageNumObjects
obj	Object

If the label of the object does not exist in the object labels of the dataset, the label will be added to the object labels of the dataset.

◆ SetImageObject() [2/3]

void eVision.EasyDeepLearning.EDataset.SetImageObject	(	int	imageIndex,
		int	objectIndex,
		eVision.ERectangleRegion	region )

Sets the specified object for the given image.

Parameters

imageIndex	Index of the image
objectIndex	Index of the object between 0 and EDataset.GetImageNumObjects
region	New region for the object

If the label of the object does not exist in the object labels of the dataset, the label will be added to the object labels of the dataset.

◆ SetImageObject() [3/3]

void eVision.EasyDeepLearning.EDataset.SetImageObject	(	int	imageIndex,
		int	objectIndex,
		string	label )

Sets the specified object for the given image.

Parameters

imageIndex	Index of the image
objectIndex	Index of the object between 0 and EDataset.GetImageNumObjects
label	New label for the object

If the label of the object does not exist in the object labels of the dataset, the label will be added to the object labels of the dataset.

◆ SetLabel()

void eVision.EasyDeepLearning.EDataset.SetLabel	(	int	index,
		string	label )

Sets the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1). This operation does not add a new label to the dataset but simply renames an existing label.

Parameters

index	Label index
label	Replacement label

◆ SetLabelColor()

void eVision.EasyDeepLearning.EDataset.SetLabelColor	(	int	index,
		eVision.ERGBColor	color )

Sets the label color for a classification label.

Parameters

index	Label index (0 to EDataset.NumLabels - 1).
color	New color for the label

◆ SetLabelOpacity()

void eVision.EasyDeepLearning.EDataset.SetLabelOpacity	(	int	index,
		byte	opacity )

Sets the label opacity for a classification label.

Parameters

index	Label index (0 to EDataset.NumLabels - 1).
opacity	New opacity for the label

◆ SetLabelWeight()

void eVision.EasyDeepLearning.EDataset.SetLabelWeight	(	int	index,
		float	weight )

Sets the weight associated to the i-th label of the dataset (starting from 0 to EDataset.NumLabels - 1). This operation does not add a new label.

Parameters

index	Label index
weight	Weight to be set

◆ SetMask()

void eVision.EasyDeepLearning.EDataset.SetMask	(	int	imageIndex,
		eVision.ERegion	mask )

Sets the mask region/don't care area for the given image. The mask is defined with respect to the region of interest of the image.

Parameters

imageIndex	Index of the image for which to get the mask region
mask	Mask to set on the image

◆ SetObjectLabel() [1/2]

void eVision.EasyDeepLearning.EDataset.SetObjectLabel	(	int	labelIndex,
		string	newLabel )

Sets an object label.

Parameters

labelIndex	Index of the object label
newLabel	New label to be set

◆ SetObjectLabel() [2/2]

void eVision.EasyDeepLearning.EDataset.SetObjectLabel	(	string	oldLabel,
		string	newLabel )

Sets an object label.

Parameters

oldLabel	Old label to change
newLabel	New label to be set

◆ SetObjectLabelColor()

void eVision.EasyDeepLearning.EDataset.SetObjectLabelColor	(	int	index,
		eVision.ERGBColor	color )

Sets the label color for an object label.

Parameters

index	Label index (0 to EDataset.NumLabels - 1).
color	New color for the label

◆ SetObjectLabelOpacity()

void eVision.EasyDeepLearning.EDataset.SetObjectLabelOpacity	(	int	index,
		byte	opacity )

Sets the label opacity for an object label.

Parameters

index	Label index (0 to EDataset.NumLabels - 1).
opacity	New opacity for the label

◆ SetObjectLabelWeight() [1/2]

void eVision.EasyDeepLearning.EDataset.SetObjectLabelWeight	(	int	labelIndex,
		float	weight )

Sets the weight of an object label.

Parameters

labelIndex	Index of the object label
weight	New weight

◆ SetObjectLabelWeight() [2/2]

void eVision.EasyDeepLearning.EDataset.SetObjectLabelWeight	(	string	label,
		float	weight )

Sets the weight of an object label.

Parameters

label	Label
weight	New weight

◆ SetOCRAnnotations()

void eVision.EasyDeepLearning.EDataset.SetOCRAnnotations	(	int	imgId,
		eVision.EasyDeepOCR.ETextLine[]	texts )

Sets the ground truth text lines for the specified image.

Parameters

imgId	Index of the image.
texts	Text lines of the image.

◆ SetRegionOfInterest()

void eVision.EasyDeepLearning.EDataset.SetRegionOfInterest	(	int	imageIndex,
		int	xOrg,
		int	yOrg,
		int	width,
		int	height )

Sets the region of interest for the specified image.

Parameters

imageIndex	Index of the image.
xOrg	ROI origin abscissa
yOrg	ROI origin ordinate
width	ROI width
height	ROI height

Any objects or segmentation located outside of the region of interest will be lost.

◆ SetSegmentationLabel()

void eVision.EasyDeepLearning.EDataset.SetSegmentationLabel	(	int	index,
		string	label )

Sets the segmentation labels.

Parameters

index	Index of the segmentation label
label	String representing the segmentation label

The segmentation label index 0 is reserved and corresponds to the "background" label. This segmentation label can't be changed.

◆ SetSegmentationLabelColor()

void eVision.EasyDeepLearning.EDataset.SetSegmentationLabelColor	(	int	index,
		eVision.ERGBColor	color )

Sets the label color for a segmentation label.

Parameters

index	Label index (0 to EDataset.NumLabels - 1).
color	New color for the label

◆ SetSegmentationLabelOpacity()

void eVision.EasyDeepLearning.EDataset.SetSegmentationLabelOpacity	(	int	index,
		byte	opacity )

Sets the label opacity for a segmentation label.

Parameters

index	Label index (0 to EDataset.NumLabels - 1).
opacity	New opacity for the label

◆ SetSegmentationLabelWeight() [1/2]

void eVision.EasyDeepLearning.EDataset.SetSegmentationLabelWeight	(	int	index,
		float	weight )

Sets the segmentation label weight.

Parameters

index	Index of the segmentation label
weight	Weight of the segmentation label

◆ SetSegmentationLabelWeight() [2/2]

void eVision.EasyDeepLearning.EDataset.SetSegmentationLabelWeight	(	string	label,
		float	weight )

Sets the segmentation label weight.

Parameters

label	Segmentation label
weight	Weight of the segmentation label

◆ SetSegmentationMap()

void eVision.EasyDeepLearning.EDataset.SetSegmentationMap	(	int	imageIndex,
		eVision.EROIBW16	segmentationMap )

Sets the segmentation map of an image. The segmentation map is a 16-bit EROIBW16 image where the value of each pixel is equal to the corresponding segmentation label index (see EDataset.GetSegmentationLabel).

If an image has no segmentation (EDataset.HasSegmentation), the getter will throw an exception.

Parameters

imageIndex	Image index
segmentationMap	Segmentation map

◆ SplitDataset() [1/3]

void eVision.EasyDeepLearning.EDataset.SplitDataset	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2 )

Splits the dataset in two parts to be used for training and validation respectively.

Parameters

d1	First part of the dataset
d2	Second part of the dataset

◆ SplitDataset() [2/3]

void eVision.EasyDeepLearning.EDataset.SplitDataset	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2,
		float	proportion )

Splits the dataset in two parts to be used for training and validation respectively.

Parameters

d1	First part of the dataset
d2	Second part of the dataset
proportion	Proportion of image of each class to put into the first part. The remaining images are put in d2

◆ SplitDataset() [3/3]

void eVision.EasyDeepLearning.EDataset.SplitDataset	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2,
		float	proportion,
		bool	random )

Splits the dataset in two parts to be used for training and validation respectively.

Parameters

d1	First part of the dataset
d2	Second part of the dataset
proportion	Proportion of image of each class to put into the first part. The remaining images are put in d2
random	Randomly sample the images.

◆ SplitDatasetForLocator() [1/3]

void eVision.EasyDeepLearning.EDataset.SplitDatasetForLocator	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2 )

Splits the dataset in two parts for training and validation of a ELocator tool. Images without object labeling are excluded from the split.

Parameters

d1	First part of the dataset
d2	Second part of the dataset

The method ensures that all the object labels are represented in d1. Thus, even when the parameter random is set to false, the images in d1 and d2 can be ordered differently than they were in the original dataset.

◆ SplitDatasetForLocator() [2/3]

void eVision.EasyDeepLearning.EDataset.SplitDatasetForLocator	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2,
		float	proportion )

Splits the dataset in two parts for training and validation of a ELocator tool. Images without object labeling are excluded from the split.

Parameters

d1	First part of the dataset
d2	Second part of the dataset
proportion	Proportion of image of each class to put into the first part. The remaining images are put in d2

The method ensures that all the object labels are represented in d1. Thus, even when the parameter random is set to false, the images in d1 and d2 can be ordered differently than they were in the original dataset.

◆ SplitDatasetForLocator() [3/3]

void eVision.EasyDeepLearning.EDataset.SplitDatasetForLocator	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2,
		float	proportion,
		bool	random )

Splits the dataset in two parts for training and validation of a ELocator tool. Images without object labeling are excluded from the split.

Parameters

d1	First part of the dataset
d2	Second part of the dataset
proportion	Proportion of image of each class to put into the first part. The remaining images are put in d2
random	Randomly sample the images.

The method ensures that all the object labels are represented in d1. Thus, even when the parameter random is set to false, the images in d1 and d2 can be ordered differently than they were in the original dataset.

◆ SplitDatasetForSegmentation() [1/3]

void eVision.EasyDeepLearning.EDataset.SplitDatasetForSegmentation	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2 )

Splits the dataset in two parts for a supervised segmenter. The two parts are to be used for training and validation respectively.

Parameters

d1	First part of the dataset
d2	Second part of the dataset

The method ensures that all the segmentation labels are represented in d1. Thus, even when the parameter random is set to false, the images in d1 and d2 can be ordered differently than they were in the original dataset.

◆ SplitDatasetForSegmentation() [2/3]

void eVision.EasyDeepLearning.EDataset.SplitDatasetForSegmentation	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2,
		float	proportion )

Splits the dataset in two parts for a supervised segmenter. The two parts are to be used for training and validation respectively.

Parameters

d1	First part of the dataset
d2	Second part of the dataset
proportion	Proportion of image of each class to put into the first part. The remaining images are put in d2

The method ensures that all the segmentation labels are represented in d1. Thus, even when the parameter random is set to false, the images in d1 and d2 can be ordered differently than they were in the original dataset.

◆ SplitDatasetForSegmentation() [3/3]

void eVision.EasyDeepLearning.EDataset.SplitDatasetForSegmentation	(	eVision.EasyDeepLearning.EDataset	d1,
		eVision.EasyDeepLearning.EDataset	d2,
		float	proportion,
		bool	random )

Splits the dataset in two parts for a supervised segmenter. The two parts are to be used for training and validation respectively.

Parameters

d1	First part of the dataset
d2	Second part of the dataset
proportion	Proportion of image of each class to put into the first part. The remaining images are put in d2
random	Randomly sample the images.

The method ensures that all the segmentation labels are represented in d1. Thus, even when the parameter random is set to false, the images in d1 and d2 can be ordered differently than they were in the original dataset.

◆ UnsetImageLabel()

void eVision.EasyDeepLearning.EDataset.UnsetImageLabel ( int index )

Unset the label of the given image of the dataset. After this operation, the given image will have no classification labels.

Parameters

index Index of the image for which to unset the label

◆ UnsetImageObjectLabeling()

void eVision.EasyDeepLearning.EDataset.UnsetImageObjectLabeling ( int imageIndex )

Unsets the object labeling for the specified image. This sets the image as not having been labeled for object detection. This image won't be use for training with a ELocator tool.

Parameters

imageIndex Index of the image

◆ UnsetOCRAnnotations()

void eVision.EasyDeepLearning.EDataset.UnsetOCRAnnotations ( int imgId )

Unsets the OCR labeling for the specified image.

Parameters

imgId Index of the image.

◆ UnsetSegmentation()

void eVision.EasyDeepLearning.EDataset.UnsetSegmentation ( int imageIndex )

Unsets the segmentation of an image. After this operation, the image will have no segmentation.

Parameters

imageIndex Image index

Property Documentation

◆ AbsoluteMaxOverlap

float eVision.EasyDeepLearning.EDataset.AbsoluteMaxOverlap

get

Absolute maximum overlap between objects in the dataset.

◆ BasePath

string eVision.EasyDeepLearning.EDataset.BasePath

set

Sets the base path for the images specified with a relative path.

The base path is not serialized. It must be set after loading a dataset file.

◆ Channels

uint eVision.EasyDeepLearning.EDataset.Channels

get

Number of channels of the first image added to the dataset.

◆ Height

uint eVision.EasyDeepLearning.EDataset.Height

get

Height of the region of interest of the first image added to the dataset.

◆ ImagesIndexesWithNoLabel

int [] eVision.EasyDeepLearning.EDataset.ImagesIndexesWithNoLabel

get

Gets a list of index corresponding to the images with no classification labels.

◆ LabeledImagesIndexes

int [] eVision.EasyDeepLearning.EDataset.LabeledImagesIndexes

get

Gets a list of index corresponding to the images that have a classification label.

◆ MaxNumObjectPerImage

int eVision.EasyDeepLearning.EDataset.MaxNumObjectPerImage

get

Maximum number of objects for an image in the dataset. If no images has object labeling (see EDataset.HasObjectLabeling), the method returns -1.

◆ NumImageFiles

int eVision.EasyDeepLearning.EDataset.NumImageFiles

get

Number of different image files contained in the dataset.

The number of image files can be different than the number of images of the dataset. An image file can correspond to several dataset images (with different ROI).

◆ NumImages

int eVision.EasyDeepLearning.EDataset.NumImages

get

Number of images in the dataset.

◆ NumImagesWithForegroundSegments

int eVision.EasyDeepLearning.EDataset.NumImagesWithForegroundSegments

get

Number of images that have a ground truth segmentation that has foreground segments and is this not entirely composed of background pixels.

◆ NumImagesWithObjectLabeling

int eVision.EasyDeepLearning.EDataset.NumImagesWithObjectLabeling

get

Number of images in the dataset that are labeled for object detection.

◆ NumImagesWithObjects

int eVision.EasyDeepLearning.EDataset.NumImagesWithObjects

get

Number of images in the dataset that are labeled for object detection and have 1 or more objects.

◆ NumImagesWithOCR

int eVision.EasyDeepLearning.EDataset.NumImagesWithOCR

get

Number of images with OCR annotations.

◆ NumImagesWithoutForegroundSegments

int eVision.EasyDeepLearning.EDataset.NumImagesWithoutForegroundSegments

get

Number of images that have a ground truth segmentation that has no foreground segments and is thus entirely composed of background pixels.

◆ NumImagesWithoutObjectLabeling

int eVision.EasyDeepLearning.EDataset.NumImagesWithoutObjectLabeling

get

Number of images in the dataset that are not labeled for object detection.

◆ NumImagesWithoutObjects

int eVision.EasyDeepLearning.EDataset.NumImagesWithoutObjects

get

Number of images in the dataset that are labeled for object detection but have been associated with no object.

◆ NumImagesWithoutOCR

int eVision.EasyDeepLearning.EDataset.NumImagesWithoutOCR

get

Number of images without OCR annotations.

◆ NumImagesWithoutSegmentation

int eVision.EasyDeepLearning.EDataset.NumImagesWithoutSegmentation

get

Number of images that doesn't have a ground truth segmentation.

◆ NumImagesWithSegmentation

int eVision.EasyDeepLearning.EDataset.NumImagesWithSegmentation

get

Number of images that have a ground truth segmentation.

◆ NumLabeledImages

int eVision.EasyDeepLearning.EDataset.NumLabeledImages

get

Number of images that have a classification label.

◆ NumLabels

int eVision.EasyDeepLearning.EDataset.NumLabels

get

Number of labels in the dataset.

◆ NumObjectLabels

int eVision.EasyDeepLearning.EDataset.NumObjectLabels

get

Number of object labels.

◆ NumSegmentationLabels

int eVision.EasyDeepLearning.EDataset.NumSegmentationLabels

get

Number of segmentation labels. A dataset has always at least one segmentation label: "background".

◆ NumTexts

int eVision.EasyDeepLearning.EDataset.NumTexts

get

Number of annotated texts in the dataset.

◆ NumUnlabeledImages

int eVision.EasyDeepLearning.EDataset.NumUnlabeledImages

get

Number of images that doesn't have any classification label.

◆ ObjectSize

int eVision.EasyDeepLearning.EDataset.ObjectSize

getset

Object size for EasyLocate Interest point.

◆ SameLabelMaxOverlap

float eVision.EasyDeepLearning.EDataset.SameLabelMaxOverlap

get

Maximum overlap between objects with the same label in the dataset.

◆ TextDataset

eVision.EasyDeepLearning.EDataset eVision.EasyDeepLearning.EDataset.TextDataset

get

Transforms the current dataset to have only subwindow containing the actual text to read.

◆ TextLabeledImagesIndexes

int [] eVision.EasyDeepLearning.EDataset.TextLabeledImagesIndexes

get

Gets a list of index corresponding to the images that have a text only label.

◆ TextOrientationDataset

eVision.EasyDeepLearning.EDataset eVision.EasyDeepLearning.EDataset.TextOrientationDataset

get

Transforms the current text dataset to duplicate it with half image flipped, adding corresponding ground truth.

◆ Width

uint eVision.EasyDeepLearning.EDataset.Width

get

Width of the region of interest of the first image added to the dataset.

Public Member Functions

Properties

Detailed Description

Constructor & Destructor Documentation

◆ EDataset() [1/2]

◆ EDataset() [2/2]

Member Function Documentation

◆ AddImage() [1/8]

◆ AddImage() [2/8]

◆ AddImage() [3/8]

◆ AddImage() [4/8]

◆ AddImage() [5/8]

◆ AddImage() [6/8]

◆ AddImage() [7/8]

◆ AddImage() [8/8]

◆ AddImageObject() [1/2]

◆ AddImageObject() [2/2]

◆ AddImages() [1/4]

◆ AddImages() [2/4]

◆ AddImages() [3/4]

◆ AddImages() [4/4]

◆ AddImagesFromCSV() [1/5]

◆ AddImagesFromCSV() [2/5]

◆ AddImagesFromCSV() [3/5]

◆ AddImagesFromCSV() [4/5]

◆ AddImagesFromCSV() [5/5]

◆ AddLabel()

◆ AddObjectLabel() [1/2]

◆ AddObjectLabel() [2/2]

◆ AddRegionToSegment() [1/2]

◆ AddRegionToSegment() [2/2]

◆ AddSegmentationLabel() [1/2]

◆ AddSegmentationLabel() [2/2]

◆ Clear()

◆ Export() [1/12]

◆ Export() [2/12]

◆ Export() [3/12]

◆ Export() [4/12]

◆ Export() [5/12]

◆ Export() [6/12]

◆ Export() [7/12]

◆ Export() [8/12]

◆ Export() [9/12]

◆ Export() [10/12]

◆ Export() [11/12]

◆ Export() [12/12]

◆ ExtractSplit()

◆ GetAvailableImageAnnotationFormat() [1/2]

◆ GetAvailableImageAnnotationFormat() [2/2]

◆ GetImageCopy() [1/2]

◆ GetImageCopy() [2/2]

◆ GetImageLabel()

◆ GetImageNumObjects()

◆ GetImageObject()

◆ GetImageObjects()

◆ GetImagePath()

◆ GetImagesIndexesWithLabel() [1/2]

◆ GetImagesIndexesWithLabel() [2/2]

◆ GetLabel()

◆ GetLabelColor()

◆ GetLabelOpacity()

◆ GetLabelWeight()

◆ GetMask()

◆ GetNumImagesForFile()

◆ GetNumImagesForSegmentationLabel() [1/2]

◆ GetNumImagesForSegmentationLabel() [2/2]

◆ GetNumImagesWithObjectLabel() [1/2]

◆ GetNumImagesWithObjectLabel() [2/2]

◆ GetNumObjectsWithLabel() [1/2]

◆ GetNumObjectsWithLabel() [2/2]

◆ GetNumPixelsForSegmentationLabel() [1/2]

◆ GetNumPixelsForSegmentationLabel() [2/2]

◆ GetNumSegmentedBlobs() [1/3]

◆ GetNumSegmentedBlobs() [2/3]

◆ GetNumSegmentedBlobs() [3/3]

◆ GetObjectLabel()

◆ GetObjectLabelColor()

◆ GetObjectLabelOpacity()

◆ GetObjectLabelWeight() [1/2]

◆ GetObjectLabelWeight() [2/2]