期刊文献+

一种自适应多窗口的立体匹配算法 被引量:6

Adaptive Multiple Windows Stereo Matching Algorithm
在线阅读 下载PDF
导出
摘要 提出了一种新型的自适应匹配窗口的方法来解决低纹理图像在立体匹配中容易出现误配的现象。该算法采用8个相同的窗口根据图像的平滑情况往8个方向选择适当的支撑区域,且其非常适合并行处理。然后根据上述步骤得到的结果提出一种自适应的区域生长法对误配和遮挡区域进行后处理。实验结果表明,该方法无论在低纹理区域还是高纹理区域都有较好的实验结果,并且运行时间与传统的固定窗口局部方法相差不大,适合实时处理。 New kind of adaptive multiple windows stereo matching algorithm was presented,which can solve that easily appeared with error matching in textureless image in stereo matching.The algorithm adopts 8 same windows according the smooth situation of images to find appropriate support area by 8 direction,and this method is very suitable for parallel processing.Then presented a kind of adaptive region growing algorithm to solve that the obtained result by the above steps,which leaves something to be desired.Experimental results show that this method is whether in textureless regions or texture regions has greatly improved,and running time close to the traditional local fixed window method,which is suitable for real-time processing.
出处 《计算机科学》 CSCD 北大核心 2012年第B06期519-521,558,共4页 Computer Science
基金 广东省部产学研结合项目(2009B080701077)资助
关键词 立体匹配 低纹理 自适应多窗口 自适应区域生长 并行处理 实时匹配 Stereo matching; Textureless; Adaptive multiple windows; Adaptive region growing; Parallel processing; Real-time matching
作者简介 曾凡志(1985-),男,硕士,主要研究方向为计算机视觉等,E-mail:zengfanzhi512@qq.com,; 鲍苏苏 男,教授,硕士生导师,主要方向为模式识别、医学图像等。
  • 相关文献

参考文献12

  • 1Scharstein D, Szellski R. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms[J]. Int. Jour. Computer Vision, 2002,47 (1/2/3): 7-42.
  • 2Kolmogorov V, Zabih R. Computing visual correspondence with occlusions using graph cuts[C]//ICCV. 2001.
  • 3Sun Jian, Zheng Nan-ning, Shum H Y. Stereo Matching Using Belief Propagation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003,25 (7) : 787-800.
  • 4Kim J, Lee K, Choi B, et al. A dense stereo matching using two- pass dynamic programming with generalized ground control points[C]//Proc. Conf. on Computer Vision and Pattern Reeognition(CVPR 2005). 2005:1075-1082.
  • 5Mc Donnel M. Box-filtering techniques [J]. Computer Graphics and Image Processing, 1981,17 : 65-70.
  • 6Crow F. Summed-area tables for texture mapping [J]. Computer Graphics, 1984,18(3):207-212.
  • 7Mattoecia S. A locally global approach to stereo correspondence [C]//IEEE Workshop on 3D Digital Imaging and Modeling (3DIM 2009). 2009.
  • 8Gerrits M,Bekaert P. Local Stereo Matching with Segmentation- based Outlier Rejection [C]//Proc. Canadian Conf. on Computer and Robot Vision(CRV 2006). 2006:66-66.
  • 9Veksler O. Fast variable window for stereo correspondence using integral images [C]//Proc. Conf. on Computer Vision and Pattern Recognition(CVPR 2003). 2003:556-561.
  • 10Hirschmuller H, Innocent P, Garibaldi J. Real-time correlation based stereo vision with reduced border errors [J]. Int. Journ. of Computer Vision, 2002 , 47 : 1-3.

同被引文献37

引证文献6

二级引证文献26

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部