留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于递推自适应权重的快速稠密立体匹配

杨奎 赵剡 苏庆华 邓年茂

杨奎, 赵剡, 苏庆华, 等 . 基于递推自适应权重的快速稠密立体匹配[J]. 北京航空航天大学学报, 2013, 39(7): 963-967.
引用本文: 杨奎, 赵剡, 苏庆华, 等 . 基于递推自适应权重的快速稠密立体匹配[J]. 北京航空航天大学学报, 2013, 39(7): 963-967.
Yang Kui, Zhao Yan, Su Qinghua, et al. Fast recursive adaptive weight stereo matching[J]. Journal of Beijing University of Aeronautics and Astronautics, 2013, 39(7): 963-967. (in Chinese)
Citation: Yang Kui, Zhao Yan, Su Qinghua, et al. Fast recursive adaptive weight stereo matching[J]. Journal of Beijing University of Aeronautics and Astronautics, 2013, 39(7): 963-967. (in Chinese)

基于递推自适应权重的快速稠密立体匹配

基金项目: 国家自然科学基金资助项目(61233005, 61074184)
详细信息
  • 中图分类号: TP 391.4

Fast recursive adaptive weight stereo matching

  • 摘要: 针对经典自适应权重稠密立体匹配算法计算量大的问题,提出了一种递推自适应权重算法.重新定义相邻像素的权重为距离衰减因子和色彩差异函数的乘积,不相邻像素权重为相邻像素权重的累乘,色彩差异越小、距离越近的像素权重越大;证明了在新的权重定义下,一维空间的匹配代价融合可以通过两次递推完成,真实图像的匹配代价融合可以通过4次递推完成,同时给出相应递推公式;递推匹配代价融合时每个像素每一视差只做4次乘法和8次加法,计算量比窗口大小为35×35的经典自适应权重算法小约两个数量级;基于递推匹配代价融合实现了一种快速稠密立体匹配算法.使用Middlebury大学的测评集测试该算法,证明了递推自适应权重算法的速度和精度均优于经典自适应权重算法.

     

  • [1] 张广军.视觉测量[M].北京:科学出版社,2008:148-158 Zhang Guangjun.Vision measurement[M].Beijing:Science Press,2008:148-158(in Chinese)
    [2] Scharstein D,Szeliski R.A taxonomy and evaluation of dense two-frame stereo correspondence algorithms[J].International Journal of Computer Vision,2002,47(1):7-42
    [3] 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
    [4] Tomasi C,Manduchi R.Bilateral filtering for gray and color images //International Conference on Computer Vision.Bombay,India:IEEE Computer Society,1998:839-846
    [5] Hu Weidong,Zhang Kang,Sun Lifeng,et al.Virtual support window for adaptive-weight stereo matching //Visual Communications and Image Processing.Tainan:IEEE Circuits and Systems Society,2011:1-4
    [6] Kowalczuk J,Psota E T,Perez L C.Real-time stereo matching on CUDA using an iterative refinement method for adaptive support-weight correspondences[J].IEEE Transactions on Circuits and Systems for Video Technology,2013,23(1):94-104
    [7] Klaus A,Sormann M,Karner K.Segment-based stereo matching using belief propagation and a self-adapting dissimilarity measure //International Conference on Pattern Recognition.Hong-kong:International Association for Pattern Recognition,2006:15-18
    [8] Scharstein D,Szeliski R.Middlebury stereo vision .(2012-04-05) .http://vision.middlebury.edu/stereo/
    [9] Dongbo M,Jiangbo L,Do M N.A revisit to cost aggregation in stereo matching:How far can we reduce its computational redundancy //International Conference on Computer Vision.Barcelona:IEEE Computer Society,2011:1567-1574
    [10] Hirschmuller H,Scharstein D.Evaluation of stereo matching costs on Images with radiometric differences[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(9):1582-1599
    [11] Yang Q.A non-local cost aggregation method for stereo matching //IEEE Conference on Computer Vision and Pattern Recognition.Providence,Rhode Island:IEEE Computer Society,2012:1402-1409
    [12] Perreault S,Hebert P.Median filtering in constant time[J].IEEE Transactions on Image Processing,2007,16(9):2389-2394
  • 加载中
计量
  • 文章访问数:  1580
  • HTML全文浏览量:  119
  • PDF下载量:  601
  • 被引次数: 0
出版历程
  • 收稿日期:  2012-08-02
  • 网络出版日期:  2013-07-30

目录

    /

    返回文章
    返回
    常见问答