基于改进Census变换的多特性立体匹配算法

计算机工程与科学

基于改进Census变换的多特性立体匹配算法

欧永东，谢小鹏

（华南理工大学机械与汽车工程学院,广东广州 510640）

收稿日期:2019-11-11 修回日期:2020-01-03 出版日期:2020-06-25 发布日期:2020-06-25

A multi-feature stereo matching algorithm

based on improved Census transform

OU Yong-dong，XIE Xiao-peng

（School of Mechnical & Automotive Engineering,South China University of Technology,Guangzhou 510640,China）

Received:2019-11-11 Revised:2020-01-03 Online:2020-06-25 Published:2020-06-25

摘要/Abstract

摘要：

针对当前立体匹配算法存在的匹配准确率低，难以达到实用的高精度水平的问题，提出了一种基于改良的Census变换与色彩信息和梯度测度相结合的多特性立体匹配算法，实现高精度的双目立体匹配。算法首先在初始代价匹配阶段，将改进的Census变换、色彩和梯度测度赋权求和得出可靠的初始匹配代价；在聚合阶段，采取高效快捷的最小生成树聚合，获得匹配代价矩阵；最后根据胜者为王法则得到初始视差图，并引入左右一致性检测等策略优化视差图，获得高精度的视差图，实验阶段对源自Middlebury上的标准测试图进行测试验证，实验结果表明，经本文算法处理得到的15组测试数据集的视差图在非遮挡区域的平均误匹配率为6.81%，算法实时响应性优良。

关键词: Census变换, 多特性, 立体匹配, 最小生成树, 视差图

Abstract:

Aiming at the problem that the low matching accuracy of the current stereo matching algorithm cannot achieve the practical high precision level. This paper proposes a multi-feature stereo matching algorithm combining improved Census transform, color information and gradient measure, so as to realize high-precision binocular stereo matching. Firstly, in the initial cost matching phase, the improved Census transform, color and gradient measure are summed to obtain a reliable initial matching cost. In the aggregation phase, the efficient and fast minimum spanning tree aggregation is adopted to obtain the matching cost matrix. Finally, the initial disparity map is obtained according to the winner's strategy, and the disparity map is optimized by the left and right consistency detection method to obtain a high-precision disparity map. Experiments on the standard test charts provided on the Middlebury website show that the average mismatch rate of the disparity maps of the 15 test data sets obtained by the algorithm is 6.81%. The algorithm has excellent real-time responsiveness.

Key words: Census transform, multiple features, stereo matching；minimum spanning tree；disparity map

欧永东，谢小鹏. 基于改进Census变换的多特性立体匹配算法[J]. 计算机工程与科学.

OU Yong-dong，XIE Xiao-peng.

A multi-feature stereo matching algorithm

based on improved Census transform

[J]. Computer Engineering & Science.

[1]	桑海伟1,3，徐孩2，熊伟程1，左羽1，赵勇1,2. 基于像素注意力的双通道立体匹配网络[J]. 计算机工程与科学, 2020, 42(05): 877-883.
[2]	柴钰，曹小京，刘杰. 基于改进绝对差值代价和动态窗口的立体匹配算法[J]. 计算机工程与科学, 2019, 41(10): 1809-1815.
[3]	郭鑫1,2，王延杰1，付东辉1,2，樊博1,2. 三种测度融合的加权引导滤波匹配算法[J]. 计算机工程与科学, 2019, 41(06): 1044-1049.
[4]	王召月，陈丽芳. 基于mean-shift全局立体匹配方法[J]. 计算机工程与科学, 2017, 39(07): 1333-1337.
[5]	王刚，李朝锋. 基于视差图和复数轮廓波变换的无参考立体图像质量评价[J]. 计算机工程与科学, 2017, 39(03): 511-518.
[6]	胡涛，朱黎，郑明辉，毛雷. 基于视差和SIFT的双目视觉移动目标识别和追踪的研究[J]. J4, 2015, 37(10): 1947-1951.
[7]	杨书新,谭伟,魏朝奇. 一种基于图结构分解的图近似查询方法研究[J]. J4, 2015, 37(02): 359-364.
[8]	焦贤龙，王晓东，周兴铭. 移动自组网中一种网络生存时间最优的广播算法[J]. J4, 2011, 33(1): 12-19.
[9]	宋汉辰吴玲达杨冰蒋杰. 校正图像对视差图计算的一种简化方法[J]. J4, 2007, 29(12): 64-66.
[10]	刘建伟卢建朱张彦军. 一个改进的调配算法[J]. J4, 2007, 29(1): 73-75.
[11]	胡庭波吴涛贺汉根. 基于立体匹配技术的光流场计算方法[J]. J4, 2006, 28(10): 50-53.