摘要
为了解决计算机视觉中对应点的误匹配问题,提出了一种在能量最小化框架下利用自适应窗口的不连续保护立体匹配算法。该算法首先引入了融合亮度和梯度信息的非相似度,然后,根据窗口内的平均匹配误差、误差方差及较大窗口的偏向误差组成的窗口代价函数,选取每个视差对应的最佳窗口作为匹配基元,并利用扫描线优化方法及深度不连续约束法寻找水平方向上的最优路径;最后,通过回溯最佳路径获得最终的稠密视差图。实验结果表明,所提算法不仅能够保留自适应窗口匹配算法的优点,较好地处理大的低纹理区域和视差不连续区域,还可加强相邻匹配基元间的视差连续性,所提算法的计算量比原有算法缩短了约3/4。
For solving the mismatching problem in computer vision, a discontinuity preserving stereo matching algorithm using variable windows in an energy minimization framework was proposed. Firstly, a dissimilarity combing intensity and gradient information was introduced. Then, according to the cost function composed of the average measurement error, error variances and the error biases of larger windows, the best window for each disparity was chosen as the matching cue, and a scanline optimization method was used to search optimum paths under the constraint of depth discontinuity. Finally, a dense disparity map was obtained by tracing optimum paths. Experimental results show that the proposed algorithm can not only process the low texture areas and depth discontinuities, but also enforce the consistency between the two neighbors. Moreover, the computation cost has almost decreased by three-fourths as compared with that of the original algorithm.
出处
《光学精密工程》
EI
CAS
CSCD
北大核心
2009年第9期2328-2335,共8页
Optics and Precision Engineering
基金
国防基础研究基金资助项目(No.J1500C002)
关键词
计算机视觉
立体匹配
非相似度
自适应窗口
扫描线优化
computer vision
stereo matching
dissimilarity
variable window
scanline optimization
