基于加权中值滤波的MST立体匹配算法

doi:10.3969/j.issn.1006-2475.2018.05.004

计算机与现代化

基于加权中值滤波的MST立体匹配算法

（贵州大学计算机科学与技术学院，贵州贵阳550025）

收稿日期:2017-11-01 出版日期:2018-06-13 发布日期:2018-06-13
作者简介: 赵大成(1991-)，男，江西九江人，贵州大学计算机科学与技术学院硕士研究生，研究方向：双目视觉与特征点匹配；通信作者：许道云（1959-），男，教授，博士生导师，研究方向：可计算性与计算复杂性，分析中的可计算性，算法设计与分析。
基金资助:
国家自然科学基金资助项目(61762019,61462001)

MST Stereo Matching Algorithm Based on Weighted Median Filtering

（College of Computer Science and Technology, Guizhou University, Guiyang 550025, China）

Received:2017-11-01 Online:2018-06-13 Published:2018-06-13

摘要/Abstract

摘要： 传统的最小生成树立体匹配算法对低纹理区域和遮挡区域不敏感，虽然最小生成树立体匹配算法后处理的中值滤波能够消除噪点，但是不能够消除边缘模糊。本文提出一种改进算法来克服这些局限性。首先，由于最小生成树匹配成本区分度不够高，研究并提出新最小生成树的匹配成本，使其可以减小不敏感区域的误匹配。其次，在后处理中使用加权中值滤波，以改善深度图像边缘。实验结果表明，在最小生成树立体匹配算法中使用改进匹配成本算法和加权中值滤波算法，在Middlebury数据集中平均误匹配率达到6.9%，本文算法在Middlebury和KITTI场景中都优于最小生成树立体匹配算法。

关键词: 最小生成树, 低纹理, 遮挡区域, 中值滤波, 匹配成本, 加权中值滤波

Abstract: The traditional Minimum Spanning Tree(MST) stereo matching algorithm is insensitive to exture-less regions and occlusion areas. Moreover, the median filter of the minimum spanning tree stereo matching algorithm can eliminate the noise, but it cannot eliminate the edge blur. This paper presents an improved algorithm to overcome these limitations. Firstly, because the cost discrimination of the minimum spanning tree matching is not enough high, the matching cost of the new minimum spanning tree is studied and it can reduce the mismatching of the insensitive region. Secondly, weighted median filtering is used in post-processing to improve image boundaries. The experimental results show that the improved matching algorithm and the weighted median filtering algorithm are used in the minimum spanning tree stereo matching algorithm, and the average false matching rate is 6.9% in the Middlebury data set. The algorithm is superior to minimum spanning tree in Middlebury and KITTI.

Key words: minimum spanning tree, exture-less, occlusion areas, median filtering, match cost, weighted median filter

中图分类号:

TP242.6

赵大成，许道云. 基于加权中值滤波的MST立体匹配算法[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2018.05.004.

ZHAO Da-cheng, XU Dao-yun. MST Stereo Matching Algorithm Based on Weighted Median Filtering[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2018.05.004.

参考文献

［1］     Scharstein D, Szeliski R. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms［J］. International Journal of Computer Vision, 2002,47(1-3):7-42. 
［2］ Boykov Y, Veksler O, Zabih R. Fast approximate energy minimization via graph cuts［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2002,23(11):1222-1239.
［3］    Kolmogorov V, Zabih R. What energy functions can be minimized via graph cuts?〖KG-*3〗［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2004,26(2):147-159. 
［4］ Yang Qingxiong, Wang L, Ahuja N. A constant-space belief propagation algorithm for stereo matching［C］// IEEE Conference on Computer Vision and Pattern Recognition. 2010:1458-1465.
［5］ Zhang Ke, Lu Jiangbo, Lafruit G. Cross-based local stereo matching using orthogonal integral images［J］. IEEE Transactions on Circuits & Systems for Video Technology, 2009,19(7):1073-1079.
［6］ Yoon K J, Kweon I S. Adaptive support-weight approach for correspondence search［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2006,28(4):650-656.
［7］   Hirschmüller H, Innocent P R, Garibaldi J. Real-time correlation-based stereo vision with reduced border errors［J］. International Journal of Computer Vision, 2002,47(1-3):229-246.
［8］ Yang Qingxiong. A non-local cost aggregation method for stereo matching［C］// IEEE Conference on Computer Vision and Pattern Recognition. 2012:1402-1409.
［9］ Ma Ziyang, He Kaiming, Wei Yichen, et al. Constant time weighted median filtering for stereo matching and beyond［C］// IEEE International Conference on Computer Vision. 2013:49-56.
［10］Li Linchen, Yu Xin, Zhang Shunli, et al. 3D cost aggregation with multiple minimum spanning trees for stereo matching［J］. Applied Optics, 2017,56(12):3411-3420.
［11］Bleyer M, Rhemann C, Rother C. PatchMatch stereo-stereo matching with slanted support windows［C］// British Machine Vision Conference. 2011:14.1-14.11.
［12］Hamzah R A, Ibrahim H, Hassan A H A. Stereo matching algorithm based on per pixel difference adjustment, iterative guided filter and graph segmentation［J］. Journal of Visual Communication & Image Representation, 2016,42(1):145-160.
［13］Perreault S, Hebert P. Median filtering in constant time［J］. IEEE Transactions on Image Processing, 2007,16(9):2389-2394.
［14］Rhemann C, Hosni A, Bleyer M, et al. Fast cost-volume filtering for visual correspondence and beyond［C］// IEEE Conference on Computer Vision and Pattern Recognition. 2011:3017-3024.

[1]	张迪1，黄倩1，陈斯斯2. 基于运动矢量投影的帧率转换算法[J]. 计算机与现代化, 2019, 0(05): 80-.
[2]	孔倩倩，赵辽英，张莉. 基于图像轮廓分析的室内窗户检测[J]. 计算机与现代化, 2018, 0(04): 56-.
[3]	叶青1,2，胡昌标1,2. 一种改进的基于图论的图像分割方法[J]. 计算机与现代化, 2016, 0(9): 64-67.
[4]	孙小军. 模糊权值网络最小生成树问题的矩阵算法[J]. 计算机与现代化, 2016, 0(9): 21-24.
[5]	韩丽娜. 基于噪声分析的椒盐噪声降噪处理方法[J]. 计算机与现代化, 2014, 0(2): 95-97.
[6]	杨晶;王元庆. 实时高速实现改进型中值滤波算法[J]. 计算机与现代化, 2013, 1(8): 30-32,4.
[7]	刘圣晓;. 基于图像处理的药片实时检测及处理系统的研究及其应用[J]. 计算机与现代化, 2013, 1(5): 66-69.
[8]	程远;. 网络最小生成树更新策略[J]. 计算机与现代化, 2012, 1(6): 125-130.
[9]	曾宪佑;黄佐华;周进朝;张勇. 一种去除椒盐噪声的自适应极值中值滤波算法[J]. 计算机与现代化, 2012, 1(11): 70-73+7.
[10]	袁杰;朱斐. 基于中值滤波和梯度锐化的边缘检测[J]. 计算机与现代化, 2012, 1(10): 89-91,1.
[11]	刘高生;马小三;王培珍. 基于改进的中值滤波和数学形态学的图像边缘检测[J]. 计算机与现代化, 2011, 1(8): 57-59,6.
[12]	叶青;唐鹏举. 一种改进的基于MST的聚类算法[J]. 计算机与现代化, 2011, 1(8): 17-19,2.
[13]	张毛女;柳薇. 一种基于边缘检测的改进的中值滤波去噪方法[J]. 计算机与现代化, 2011, 1(3): 63-65,7.
[14]	耿帅;王希常. 一种基于中值滤波的图像去噪算法[J]. 计算机与现代化, 2011, 1(11): 90-3.
[15]	冉玉梅;王洪国;杨玉会;许镇. 基于自适应局部阈值的彩色图像分割[J]. 计算机与现代化, 2010, 1(01): 109-112.

基于加权中值滤波的MST立体匹配算法

MST Stereo Matching Algorithm Based on Weighted Median Filtering

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价