ROBOT2016001-007多种相似性测度结合的局部立体匹配算法卢迪.pdf

第38 卷第1 期 机器人 ROBOT Vol.38, No.1 2016 年1 月 Jan., 2016 DOI ：10.13973/ki.robot.2016.0001 多种相似性测度结合的局部立体匹配算法 卢 迪，林 雪 （哈尔滨理工大学电气与电子工程学院，黑龙江 哈尔滨 150080） 摘 要：针对立体匹配中匹配代价和支持窗口难以选择的问题，提出一种将多种相似性测度相结合的局部立 体匹配算法．首先，构造匹配代价，结合图像的Census 变换、WLD （Weber 局部描述符）特征、图像色彩信息以 及图像梯度信息作为匹配代价；然后，使用引导滤波器对匹配代价进行聚合；最后，针对WTA （赢者全取）策略 引入的视差选择歧义问题和左右一致性检测（LRC ）引入的水平条纹问题，提出了一种基于可信度和加权滤波的 视差修正算法．利用Middlebury 测试平台提供的标准测试图像对本文算法进行测试，其平均错误匹配率为5.30%， 与FastBilateral 算法等一些公认的性能优异算法相比，本文算法提高了匹配准确率． 关键词：局部立体匹配；匹配代价；引导滤波；视差修正 中图分类号：TP391.4 文献标识码：A 文章编号：1002-0446(2016)-01-0001-07 A Local Stereo Matching Algorithm Based on the Combination of Multiple Similarity Measures LU Di ，LIN Xue (School of Electrical and Electronic Engineering, Harbin University of Science and Technology, Harbin 150080, China) Abstract: Aiming at the difficulties in choosing matching cost and support window in stereo matching, a local stereo matching algorithm based on the combination of multiple similarity measures is proposed. Firstly, a matching cost is con- structed, in which the census transform of image, the WLD (Weber local descriptor) feature of image, the color information of image and the gradient information of image are combined. Secondly, the guided filter is used to aggregate matching cost. Finally, a disparity refinement algorithm based on confidence and weighted filtering is proposed to eliminate the disparity choosing ambiguity brought by WTA (winner take all) strategy and horizontal stripe brought by LRC (left-right consistency) check. The standard test images