A New Fast and Robust Stereo Matching Algorithm for Robotic Systems

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