摘要
基于马尔可夫随机场(MRF)的立体匹配算法利用MRF模型来对匹配取值进行连续性约束。然而,MRF模型是产生式模型,图像自身特征难以得到准确描述。提出了一种基于图像分割的立体匹配算法SGC。SGC算法预先对图像进行分割,基于图像分割信息建立立体匹配的MRF模型,从而连续性(平滑)约束可以保留视差图中分割的边缘信息;并针对图像的深度连续性约束,定义了一个反映图像自身特征的新能量函数,应用于图割算法,提高了视差计算精度。实验结果表明,与以往算法相比,SGC算法更准确地反映了图像中深度信息,避免了平滑约束所引入的误差,有效提高了视差计算精度。
The stereo matching algorithm based on Markov Random Field(MRF) restricts the continuity of the disparity by the MRF model, but it can not describe the image feature exactly due to the generative property of the model. This paper presented a stereo matching algorithm of SGC based on image segmentation. The SGC algorithm built the MRF model by using the result of image segmentation; thereby, the edge information of the disparity map could be kept in the continuity (smoothness) constraints. Moreover, to improve the disparity accuracy, a new energy function was well designed to restrict the depth-continuity of an image and applied to the Graph Cut (GC) algorithm to describe the image feature. The experimental results show that this SGC algorithm can reflect the depth information more exactly than the existing algorithm and achieve a high-precision disparity by avoiding the error resulting from the continuity constraints.
出处
《计算机应用》
CSCD
北大核心
2011年第1期175-178,193,共5页
journal of Computer Applications
基金
国家973计划项目(2009CB723803)
关键词
马尔可夫随机场
图割算法
立体匹配
图像分割
视差图
Markov Random Field (MRF)
Graph Cut (GC) algorithm
stereo match
image segmentation
disparity map