一种基于自适应权重SAD与Census融合的匹配算法
|
摘要
立体匹配算法中的SAD算法与Census算法较为常用,本文分析了两种算法的特点,在改进SAD算法后,将其与Census算法融合,提出了自适应权重的SAD-Census算法。试验表明,与SAD算法和Census算法相比,该算法匹配效果更好,对实际场景拍摄图像的匹配也有较好的稳健性。
The SAD algorithm and Census algorithm are commonly used in stereo matching. In this paper,the characteristics of the two methods are analysed. After the SAD algorithm is improved,and it is fused with Census to propose SAD-Census algorithm with adaptive weights. The experiments show that,compared with traditional SAD algorithm and Census algorithm,the proposed algorithm is more efficient and robust to the matching of actual scene images.
The SAD algorithm and Census algorithm are commonly used in stereo matching. In this paper,the characteristics of the two methods are analysed. After the SAD algorithm is improved,and it is fused with Census to propose SAD-Census algorithm with adaptive weights. The experiments show that,compared with traditional SAD algorithm and Census algorithm,the proposed algorithm is more efficient and robust to the matching of actual scene images.
引文
[1]靳建立,范永弘,纪松.基于面特征的整体影像匹配方法[J].测绘通报,2008(9):13-15.
[2]陈爱华,高诚辉,何炳蔚.立体视觉中的图像立体匹配方法[J].中国工程机械学报,2014,12(3):194-198.
[3]耿英楠.立体匹配技术的研究[D].长春:吉林大学,2014.
[4]岳陈平,孟丽娅.基于算法的立体匹配的实现[J].微型机与应用,2013,32(6):41-43.
[5]王军政,朱华健,李静.一种基于Census变换的可变权值立体匹配算法[J].北京理工大学学报,2013,33(7):704-710.
[6]刘正.计算机视觉中立体匹配技术的研究[D].南京:南京理工大学,2005.
[7]张来刚,魏仲慧,何昕,等.面向低纹理图像的快速立体匹配[J].液晶与显示,2013,28(3):450-458.
[8]张浩峰,赵春霞.一种利用动态规划和左右一致性的立体匹配算法[J].中国图象图形学报,2008,13(4):756-760.
[9] HUMENBERGER M,ZINNER C,WEBER M,et al.A Fast Stereo Matching Algorithm Suitable for Embedded Real-time Systems[J]. Computer Vision and Image Understanding,2010,114(11):1180-1202.
[10]曹之乐,严中红,王洪.双目立体视觉匹配技术综述[J].重庆理工大学学报(自然科学版),2015,29(2):70-75.
[11] SCHATEIN 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.
[12]张来刚,魏仲慧,何昕,等.面向低纹理图像的快速立体匹配[J].液晶与显示,2013,28(3):450-458.
[13]桑瑞娟,王姮,张华,等.一种改进的区域双目立体匹配方法[J].传感器与微系统,2012,31(8):57-59.
[14]张华东,潘晨,章东平.基于视觉显著性的区域立体匹配算法[J].计算机应用,2015,35(12):3565-3569.
[15]郑楷鹏.摄像机标定及立体匹配技术研究[D].南京:南京理工大学,2017.
[2]陈爱华,高诚辉,何炳蔚.立体视觉中的图像立体匹配方法[J].中国工程机械学报,2014,12(3):194-198.
[3]耿英楠.立体匹配技术的研究[D].长春:吉林大学,2014.
[4]岳陈平,孟丽娅.基于算法的立体匹配的实现[J].微型机与应用,2013,32(6):41-43.
[5]王军政,朱华健,李静.一种基于Census变换的可变权值立体匹配算法[J].北京理工大学学报,2013,33(7):704-710.
[6]刘正.计算机视觉中立体匹配技术的研究[D].南京:南京理工大学,2005.
[7]张来刚,魏仲慧,何昕,等.面向低纹理图像的快速立体匹配[J].液晶与显示,2013,28(3):450-458.
[8]张浩峰,赵春霞.一种利用动态规划和左右一致性的立体匹配算法[J].中国图象图形学报,2008,13(4):756-760.
[9] HUMENBERGER M,ZINNER C,WEBER M,et al.A Fast Stereo Matching Algorithm Suitable for Embedded Real-time Systems[J]. Computer Vision and Image Understanding,2010,114(11):1180-1202.
[10]曹之乐,严中红,王洪.双目立体视觉匹配技术综述[J].重庆理工大学学报(自然科学版),2015,29(2):70-75.
[11] SCHATEIN 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.
[12]张来刚,魏仲慧,何昕,等.面向低纹理图像的快速立体匹配[J].液晶与显示,2013,28(3):450-458.
[13]桑瑞娟,王姮,张华,等.一种改进的区域双目立体匹配方法[J].传感器与微系统,2012,31(8):57-59.
[14]张华东,潘晨,章东平.基于视觉显著性的区域立体匹配算法[J].计算机应用,2015,35(12):3565-3569.
[15]郑楷鹏.摄像机标定及立体匹配技术研究[D].南京:南京理工大学,2017.