Enhancing stereo matching with varying illumination through histogram information and normalized cross correlation
Mohammed Baydoun, Mohamad Adnan Al‐Alaoui · 2013
Stereo matching is considered with different local approaches. The problem is addressed while aiming at reducing the effect of varying colors of pixel pairs that is usually due to varying illumination, digitization, camera, noise, or others. This is done by utilizing histogram information of the stereo pair in an efficient manner that does not involve high processing. Also, the illumination invariant Normalized Cross Correlation is used to restore the pair to better conditions and enhance the matching.