A New Fast and Robust Stereo Matching Algorithm for Robotic Systems

In this paper, we propose a new area-based stereo matching method by improving the classical Census transform. It is a difficult task to match corresponding points in two images taken by stereo cameras, mostly under variant illumination and non-ideal conditions. The classic Census nonparametric transform did some improvements in the accuracy of disparity map in these conditions but it also has some disadvantages. The results were not robust under different illumination, and because of the complexity the performance was not suitable for real-time robotic systems. In order to solve these problems, this paper presents an improved Census transform using Maximum intensity differences of the pixel placed in the center of a defined window and the pixels in the neighborhood to reduce complexity and obtain better performance and needs a smaller window size to obtain best accuracy compared to the Census transform. Experimental results show that the proposed method, achieves better efficiency in term of speed and robustness against illumination changes.