分层正交动态规划立体匹配算法

A hierarchical stereo algorithm using orthogonal dynamic programming

提出一种分层正交动态规划立体匹配算法.首先利用正交动态规划算法在低分辨率图像的DSI(disparity-space image)视差空间图中进行初匹配,然后在Delta DSI(delta disparity-space image)视差变化空间图中进行精匹配,从而获得高分辨率图像的视差图.实验结果表明,该算法不仅可以改善传统动态规划算法产生的带状条纹瑕疵,而且匹配率高、速度快、应用范围广.

A hierarchical stereo algorithm using orthogonal dynamic programming was proposed. First, the coarse matching was processed in DSI （disparity-space image） of low-resolution images using orthogonal dynamic programming. Then, the fine matching was processed in Delta DSI to get the high-precision an, dense disparity map. Experimental results show that the algorithm can alleviate the effect of horizonta streaks caused by traditional DP-based stereo matching algorithm; moreover, it has a high matching-rate a fast speed and a wide range of applicatiions.