Stereo matching method based on hybrid matching cost and adaptive window

Abstract

The invention provides a stereo matching method based on the hybrid matching cost and an adaptive window. Firstly, as an improvement on the traditional gradient matching cost, gradient phase information is introduced, wherein to further improve the robustness of the algorithm, the new hybrid matching cost is provided by normalizing the gray level SAD cost through an exponential function. Then, the adaptive window can be built based on the criss-cross adaptive window generation method according to the color and spatial position relation between adjacent pixels, wherein a large window is provided for a low-texture area to improve matching precision, a small window is generated in a high-texture area to protect detail information of the edge of an object and the like, and the aggregated cost is selected according to the Winner-Takes-All strategy so that a disparity value corresponding to the minimum value of the total cost can serve as an initial matching result. Finally, a disparity refinement method based on a local disparity histogram is provided, and a high-precision disparity map is obtained. As is shown in experimental results, the stereo matching method is high in machining precision and has high robustness under the condition of amplitude distortion.

Description

Technical field: [0001] The present invention relates to a method for stereo matching of binocular images to obtain depth maps, in particular to a stereo matching method based on hybrid matching costs and adaptive windows. The depth map results obtained by the present invention can be further applied to visual measurement and three-dimensional reconstruction , virtual viewpoint image rendering and other fields. Background technique: [0002] Stereo matching is a classic problem in computer vision and has always been a research hotspot. Over the years, researchers have proposed a large number of algorithms to try to solve this problem, but due to the pathological nature of the problem itself, there is no perfect solution yet. Some typical Stereo Matching Algorithm gives a comprehensive overview of various main methods. They summarized the stereo matching process into four steps: matching cost calculation, matching cost aggregation, initial disparity calculation and dispari...

AI Technical Summary

Problems solved by technology

[0004] The traditional fixed window has the problem of changing the size of the window: if the window is too small, the matching cost will be too low, and error matching will easily occur in low-texture areas; if the window is too...

Images