中文摘要：

针对双目视觉立体匹配中的视差优化问题, 提出一种基于稳定树形结构的视差优化算法. 在双目匹配问题中, 视差可以通过检测左右眼2 张视觉成像图片中的对应点的信息来计算得出, 以生成三维深度图像, 继而通过视差优化这一步骤提高三维深度图像的质量. 从计算视差中支持域的角度出发, 用稳定度的概念来衡量支持域的特征;通过基于稳定度的树结构来评估和重构支持域, 用于之后的代价聚合, 以减少视差错误. 除了室内图片, 文中方法还被拓展到了真实路面的数据集, 其在移除大块视差错误和整合碎片上取得了明显优于其他方法的效果; 与传统的基于树结构的方法相比, 在保持精确度的同时降低了70%的聚合时间, 极大地提高了视差优化的速度.

英文摘要：

In this paper, a disparity refinement method with stability-based tree for stereo matching is proposed.In stereo matching, disparity information can be extracted by examining the relative positions of objects in thetwo panels. We focus on refinement and let this step play an effective role in improving the quality of the disparitymap. We observe the features of support regions which produce disparity errors and summarize the reason asunstableness. By developing stability-based tree to evaluate and reconstruct support regions for cost aggregation,the proposed method achieves effective performance in removing outliers. Extensive experiments on both laboratoryand re-al-world road datasets demonstrates that the proposed method outperforms existing algorithms in removinglarge error parts as well as smoothing fractions. It reduces more than 70% aggregation time comparedwith traditional tree method without loss of accuracy.