Color stereo matching based on self-organization neural networks
Xijun Hua, Masahiro Yokomichi, Michio Kono · 2004
Stereo matching is the key issue of stereo vision. In the literature, most of the stereo matching algorithms have been limited to gray level images. In this paper, we propose a new color stereo matching approach based on self-organization neural networks. For the real images, we propose a segmentation method to deal with the initial similarity map. The final similarity map is established by taking logical AND calculation of different color feature similarities so as to make full use of the color information. Experimental results have shown that the quality of the stereo matching can be considerably improved by using appropriate color matching algorithm comparing with the conventional gray value algorithm.