EasyHDRFusion

Support

Required licenses

EasyImage

Recommended images

Any image from the folder Sample Images/EasyImage/HDR

Location

Image Processing/EasyHDRFusion

Purpose

This sample program demonstrates how to:

  • Perform an HDR Fusion on a specific set of images.

The process steps are:

  1. Load a set of color images.

  2. (Optionally) Set the related exposure time for each image.

  3. (Optionally) Set an alternative merger algorithm or an alternative tone-mapper algorithm.

  4. Perform the HDR Fusion.

Code highlights

  • Perform the HDR Fusion from a set of images.

    std::vector<EImageC24> imgs; // previously loaded, all images have the same size
std::vector<float> exposureTimes; // previously loaded

EImageC24 output(imgs.back()); // its dimension must previously set, same size as the input images

EHDRFusion fusion; // by default, the fuser uses a merger that uses the Mertens algorithm, and a linear tone-mapper

  1. Set the related exposure time for each image.

Attention

If a merger that is not of type Mertens or a tone-mapper is used, the related exposure time but be set for each image:

```c++
std::vector<float> times = {
  1.0f / 0.03125f,
  1.0f / 0.0625f,
  1.0f / 0.125f, 1.0f / 0.25f,
  1.0f / 0.5f,  1.0f / 1.0f,
  1.0f / 2.0f,
  1.0f / 4.0f,
  1.0f / 8.0f,
  1.0f / 16.0f,
  1.0f / 32.0f,
  1.0f / 64.0f,
  1.0f / 128.0f,
  1.0f / 256.0f,
  1.0f / 512.0f,
  1.0f / 1024.0f}; // must have the same number of elements as the input `imgs`
```

  1. Set an alternative merger algorithm or an alternative tone-mapper algorithm.

The type of merger or tone-mapper can be set at the creation of the creation of the HDRFusion object:

```c++
EDebevecMergerParameters mergerParams; // can also be ERobertsonMergerParameters or EMertensMergerParameters
EReinhardToneMapperParameters toneMapperParams; // can also be EDragoToneMapperParameters or ELinearToneMapperParameters or EMantiukToneMapperParameters

EHDRFusion fusion(mergerParams, toneMapperParams);
```

  1. Perform the HDR Fusion.

    EImageC24 output;
fusion.Fuse(imgs, output); // if using Mertens merger algorithm
// OR
fusion.Fuse(imgs, output, times); // if not using Mertens merger algorithm