液晶与显示, 2018, 33 (4): 357, 网络出版: 2018-08-28   

基于改进稳态匹配概率的立体匹配算法研究

Stereo matching algorithm based on improved steady-state matching probability
作者单位
长春理工大学 电子信息工程学院,吉林 长春 130022
摘要
针对稳态匹配概率(Steady-State Matching Probability,SSMP)立体匹配算法在处理视差范围大的测试图中产生的空洞现象以及使用该算法后由于右视差图中的错误视差导致的左视差图中正确视差丢失问题,提出一种基于稳态匹配概率和半全局匹配(Semi-Global Matching,SGM)相结合的立体匹配算法。首先使用SSMP算法求取初始视差图。接着,使用基于爬山法颜色分割的填充准则进行填充。然后使用SGM算法重新获取视差图,将两幅视差图中一致的视差信息填充到经过左右一致性检测后的含有空洞的视差图中。最后,使用SSMP算法中的空洞填充和中值滤波得到精化后的视差图。实验结果表明,改进后的SSMP算法在Middlebury测试平台上第2版本的四组图像的平均匹配误差从538%减少到523%,第3版本部分测试图像的平均匹配误差从24.7%减少到21.5%,该算法能很好地处理上述问题,有效提高匹配精确度,且具有鲁棒性。
Abstract
Steady-State Matching Probability (SSMP) stereo matching algorithm will generate hole phenomenon in the test charts with large parallax. And the correct disparity will be lost in the left disparity map due to the wrong parallax in the right disparity map after using the algorithm. In order to solve these problems, a stereo matching algorithm based on the combination of SSMP and semi-global matching (SGM) is proposed. First, the initial disparity map is obtained by using the SSMP algorithm. Next, the disparity map is filled with the filling criteria based on the Hill-climbing color segmentation. Then, the disparity map is retrieved by the SGM algorithm. The consistent disparity information in the two disparity maps is filled in the disparity map with holes after the left-right consistency detection. Finally, the refined disparity map is obtained by the empty filling and median filtering in the SSMP algorithm. The experimental results demonstrate that the average false matching ratio of the improved SSMP algorithm decreases from 5.38% to 5.23% on the second version and decreases from 24.7% to 21.5% on the third version of Middlebury testing benchmark. The proposed algorithm is able to deal with the above problems well, improve the matching accuracy effectively and have good robustness.
参考文献

[1] 王宇,朴燕.基于多视差函数拟合的集成成像深度提取方法[J].光学学报,2015,35(4):0411002.

    WANG Y, PIAO Y. Depth extraction based on function fitting for multiple disparities in integral imaging [J]. Acta Optica Sinica, 2015, 35(4): 0411002. (in Chinese)

[2] 李蓉,邓春健,邹昆.一种基于MRF的单幅图像数据的三维重构方法研究[J].液晶与显示,2016,31(3):301-309.

    LI R, DENG C J, ZOU K. 3D reconstruction method based on single image data by MRF [J]. Chinese Journal of Liquid Crystals and Displays, 2016, 31(3): 301-309. (in Chinese)

[3] FENG L T, QIN K H. Superpixel-based graph cuts for accurate stereo matching [J]. IOP Conference Series: Earth and Environmental Science, 2017, 69(1): 012161.

[4] VEKSLER O. Stereo correspondence by dynamic programming on a tree [C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, CA, USA: IEEE, 2005, 2: 384-390.

[5] 祝世平,闫利那,李政.基于改进Census变换和动态规划的立体匹配算法[J].光学学报,2016,36(4):0415001.

    ZHU S P, YAN L N, LI Z. Stereo matching algorithm based on improved Census transform and dynamic programming [J]. Acta Optica Sinica, 2016, 36(4): 0415001. (in Chinese)

[6] 许金鑫,李庆武,刘艳,等.基于色彩权值和树形动态规划的立体匹配算法[J].光学学报,2017,37(12):1215007.

    XU J X, Li Q W, LIU Y, et al. Stereo matching algorithm based on color weights and tree dynamic programming [J]. Acta Optica Sinica, 2017, 37(12): 1215007. (in Chinese)

[7] FELZENSZWALB P F, HUTTENLOCHER D P. Efficient belief propagation for early vision [J]. International Journal of Computer Vision, 2006, 70(1): 41-54.

[8] 周自维,樊继壮,赵杰,等.基于置信传播的立体匹配并行算法[J].光学 精密工程,2011,19(11):2774-2781.

    ZHOU Z W, FAN J Z, ZHAO J, et al. Parallel stereo matching algorithm base on belief propagation [J]. Optics and Precision Engineering, 2011, 19(11): 2774-2781. (in Chinese)

[9] 赵立荣,朱玮,曹永刚,等.改进的加速鲁棒特征算法在特征匹配中的应用[J].光学 精密工程,2013,21(12):3263-3271.

    ZHAO L R, ZHU W, CAO Y G, et al. Application of improved SURF algorithm to feature matching [J]. Optics and Precision Engineering, 2013, 21(12): 3263-3271. (in Chinese)

[10] 胡汉平,朱明.基于种子点传播的快速立体匹配[J].光学 精密工程,2015,23(3):887-894.

    HU H M, ZHU M. Fast stereo matching based on seed pixel propagation [J]. Optics and Precision Engineering, 2015, 23(3): 887-894. (in Chinese)

[11] MEI X, SUN X, DONG W M, et al. Segment-tree based cost aggregation for stereo matching [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Portland, OR, USA: IEEE, 2013: 313-320.

[12] MIN D B, SOHN K. Cost aggregation and occlusion handling with WLS in stereo matching [J]. IEEE Transactions on Image Processing, 2008, 17(8): 1431-1442.

[13] HOSNI A, RHEMANN C, BLEYER M, et al. Fast cost-volume filtering for visual correspondence and beyond [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(2): 504-511.

[14] YANG Q X. A non-local cost aggregation method for stereo matching [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA: IEEE, 2012: 1402-1409.

[15] HAM B, MIN D B, OH C, et al. Probability-based rendering for view synthesis [J]. IEEE Transactions on Image Processing, 2014, 23(2): 870-884.

[16] 麻祥才,王晓红,朱明,等.基于LCD显示器光谱特性的图像颜色一致性研究[J].发光学报,2017,38(5):669-674.

    MA X C, WANG X H, ZHU M, et al. Image color consistency based on spectral characteristics of LCD monitor [J]. Chinese Journal of Luminescence, 2017, 38(5): 669-674. (in Chinese)

[17] ZHAN Y L, GU Y Z, HUANG K, et al. Accurate image-guided stereo matching with efficient matching cost and disparity refinement [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2016, 26(9): 1632-1645.

[18] HIRSCHMULLER H. Stereo processing by semiglobal matching and mutual information [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(2): 328-341.

[19] 葛忠孝,邢帅,夏琴,等.基于树形结构的半全局立体匹配算法[J].计算机工程,2016,42(8):243-248,254.

    GE Z X, XING S, XIA Q, et al. Semi-global stereo matching algorithm based on tree structure [J]. Computer Engineering, 2016, 42(8): 243-248, 254. (in Chinese)

[20] GRUZMAN I S. Using gradient tensors of the second and third orders for segmentation of images containing textures with structural redundancy [J]. Optoelectronics Instrumentation and Data Processing, 2016, 52(1): 17-23.

[21] 陈璐宇,周春艳.基于分块颜色直方图和GWLBP的图像检索算法[J].液晶与显示,2017,32(9):755-763.

    CHEN L Y, ZHOU C Y. Image retrieval algorithm based on block color histogram and GWLBP [J]. Chinese Journal of Liquid Crystals and Displays, 2017, 32(9): 755-763. (in Chinese)

[22] 谢娟英,周颖,王明钊,等.聚类有效性评价新指标[J].智能系统学报,2017,12(6):873-882.

    XIE J Y, ZHOU Y, WANG M Z, et al. New criteria for evaluating the validity of clustering [J]. CAAI Transactions on Intelligent Systems, 2017, 12(6): 873-882. (in Chinese)

张建业, 朴燕. 基于改进稳态匹配概率的立体匹配算法研究[J]. 液晶与显示, 2018, 33(4): 357. ZHANG Jian-ye, PIAO Yan. Stereo matching algorithm based on improved steady-state matching probability[J]. Chinese Journal of Liquid Crystals and Displays, 2018, 33(4): 357.

本文已被 3 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!