多种相似性测度结合的局部立体匹配算法

doi:10.13973/j.cnki.robot.2016.0001

摘要
图/表
参考文献
相关文章 (1)

全文: PDF (599 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
卢迪, 林雪. 多种相似性测度结合的局部立体匹配算法[J]. 机器人, 2016, 38(1): 1-7.DOI: 10.13973/j.cnki.robot.2016.0001. LU Di, LIN Xue. A Local Stereo Matching Algorithm Based on the Combination of Multiple Similarity Measures. ROBOT, 2016, 38(1): 1-7. DOI: 10.13973/j.cnki.robot.2016.0001.

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．

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	卢迪
	林雪

关键词 ：局部立体匹配, 匹配代价, 引导滤波, 视差修正

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 constructed, 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 provided by Middlebury test platform are used to test the proposed algorithm, and the percentage of bad matching pixel is 5.30%. Comparing with some high-performance algorithms, such as FastBilateral algorithm, the proposed method can achieve a higher matching accuracy.

Key words： local stereo matching matching cost guided filter disparity refinement

收稿日期: 2015-08-13 录用日期: 2015-12-01

TP391.4

通讯作者: 林雪，lxhust123456@163.com E-mail: lxhust123456@163.com

作者简介: 卢迪（1971-），女，博士，教授.研究领域：图像信息处理，多源传感器数据融合.
林雪（1991-），女，硕士生.研究领域：双目立体视觉，图像处理.

链接本文:

https://robot.sia.cn/CN/10.13973/j.cnki.robot.2016.0001 或 https://robot.sia.cn/CN/Y2016/V38/I1/1

[1] 周龙,徐贵力,李开宇,等.无人机导航场景中立体匹配算法研究[J].航空兵器,2012(1):48-52. Zhou L, Xu G L, Li K Y, et al. Research on stereo matching in UAV navigation environment[J]. Aero Weaponry, 2012(1): 48-52.

[2] Stentoumis C, Grammatikopoulos L, Kalisperakis I, et al. A local adaptive approach for dense stereo matching in architectural scene reconstruction[J]. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2013, 40(5W1): 219-226.

[3] 赵敏,刘秉琦,武东升.增强现实装配和维修系统的技术研究[J].光学仪器,2012,34(2):16-20. Zhao M, Liu B Q, Wu D S.Study on the techniques of augmented reality installation and maintenance system[J]. Optical Instruments, 2012, 34(2): 16-20.

[4] Gauglitz S, Hoellerer T, Turk M. Evaluation of interest point detectors and feature descriptors for visual tracking[J]. International Journal of Computer Vision, 2011, 94(3): 335-360.

[5] Humenberger M, Zinner C, Weber M, et al. A fast stereo matching algorithm suitable for embedded real-time system[J]. Computer Vision and Image Understanding, 2010, 114(11): 1180-1202.

[6] Scharstein D. Matching images by comparing their gradient fields[C]//IAPR International Conference on Pattern Recognition. Piscataway, USA: IEEE, 1994: 572-575.

[7] 李静,于虹.区域的实时立体匹配的改进算法[J].西安工业大学学报,2012,32(3):231-235. Li J, Yu H. Improved stereo matching algorithm based on region[J]. Journal of Xi'an Technological University, 2012, 32(3): 231-235.

[8] Fusiello A, Roberto V, Trucco E. Efficient stereo with multiple windowing[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, USA: IEEE, 1997: 858-863.

[9] Yoon K J, Kweon I S. Adaptive support-weight approach for correspondence search[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(4): 650-656.

[10] Zhao Z S, Feng X, Teng S H, et al. Multiscale point correspondence using feature distribution and frequency domain alignment[J]. Mathematical Problems in Engineering, 2012: No. 382369.

[11] Zhao Z S, Feng X, Wei F, et al. Learning representative features for robot topological localization[J]. International Journal of Advanced Robotic System, 2013, 10: No.215.

[12] Zhao Z S, Zhang L, Zhao M, et al. Gabor face recognition by multi-channel classifier fusion of supervised kernel manifold learning[J]. Neurocomputing, 2012, 97: 398-404.

[13] Zhao Z S, Feng X, Lin Y Y, et al. Evolved neural network ensemble by multiple heterogeneous swarm intelligence[J]. Neurocomputing, 2015, 149: 29-38.

[14] Chen J, Shan S G, He C, et al. WLD: A robust local image descriptor[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(9): 1705-1720.

[15] He K M, Sun J, Tang X O. Guided image filter[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(6): 1397-1409.

[16] Crow F C. Summed-area tables for texture mapping[J]. Computer and Graphics, 1984, 18(3): 207-212.

[17] Mattoccia S, Giardino S, Gambini A. Accurate and efficient cost aggregation strategy for stereo correspondence based on approximated joint bilateral filtering[C]//Asian Conference on Computer Vision. Berlin, Germany: Springer Verlag, 2009: 371-380.