A neural network approach to color constancy (ICCAS 2011)

This thesis presents a neural network based approach to AWB. A neural network is used to estimate the chromaticity of the illuminant in a scene based only on the image data collected by a digital camera. This is accomplished by training the neural network to learn the relationship between the pixels in a scene and the chromaticity of the scene´s illumination. From a computational perspective, the goal of color constancy is defined to be the transformation of a source image taken under an unknown illuminant, to a target image, identical to one that would have been obtained by the same camera, for the same scene, under a standard illuminant. Neural networks offer better generalization and dynamic adaptations to changes in the environment because of their learning capabilities and lack of in-built constraints.