The Quality of Disparity Maps
Stereo matching algorithm, Sum of squared difference, Disparity map, RMSE, k-means segmentation
Disparity calculation based on stereo vision is an important issue in computer vision research as it has applications in several areas of image processing in general. In this dissertation, a stereo matching based on segmentation is presented. Quality of disparity map is an important and challenging problem in stereo matching. Hence, We proposed a disparity estimation method considering hybrid approach to generate improved disparity map, based on k-means and refine disparity map of image by SSD (sum of squared difference) cost function. Lastly, we applied quality measurement matrix RMSE (root mean square error) to measures the quality of disparity map.
Volume 3 Issue 2
Pages. 1013-1016