GSOC: Adding computational photography algorithms for better image quality

[MykhailoTrushch]

Note: this is a copy of an issue in from opencv/opencv: Link

As part of GSOC 2025 idea 11, implement algorithms for enhancement of technical quality of images.

Based on the conversation in opencv-gsoc-202X mailing list, the main prioritized algorithms are:

• Smarter white balance algorithms
• Color corrections
• vignetting correction, chromatic aberration correction
• Exposure, contrast, temperature correction and enhancement in color images

In my proposal, I outline some possible algorithms that would in the beginning be implemented for opencv_contrib and could be migrated to opencv, mostly in the photo module. In particular, in order of approximate importance:

1. Auto white balance: implement FFCC
2. Chromatic aberration correction: implement algorithm of Chang et al.
3. Denoising: implement a CNN-based method as a module like dnn_superres
4. Exposure correction: implement (Adaptive, Weighted) Gamma Correction
5. Superresolution: check if any additional newer models for superresolution could be added

If I am accepted as a contributor in GSOC, I expect to pursue 2-3 of these topics in-depth to fit in the timeframe of the medium-sized project (~175 hours). Also, the choice of algorithms is preliminary and might change after additional research and feedback.

More info in my proposal doc: Link to Google Docs