一种新颖的自适应权重Census变换立体匹配算法

Novel stereo matching algorithm for adaptive weight Census transform

针对当前Census变换立体匹配算法深度不连续区域匹配精度低的缺陷,提出了一种新颖的自适应权重的Census变换立体匹配算法。在Census变换阶段计算变换窗口中心点上下左右四个像素的均值,得到中心点与该均值的差的绝对值,通过判断该绝对值的大小来确定中心点灰度值;为了有区别地对待窗口内各像素点,引入自适应权重,通过线性分段型函数计算自适应权值。在代价聚合阶段同样引入自适应权重并采用变化的聚合窗口,通过聚合窗口中心点和其左右两点的梯度值来确定聚合窗口的大小。实验结果表明,算法的匹配效果优于目前的Census变换立体匹配算法,在深度不连续区域匹配效果显著改善,而且没有明显降低实时性和增加硬件实现的难度。

Aiming at the limitation of low accuracy in the region where the depth is not continuous of the current Censustransform stereo matching algorithm, a novel stereo matching algorithm for adaptive weight Census transform is proposed.An absolute value derived from the central pixel and its cross-mean is to determine the central gray valuein the stage ofCensus transform; With the purpose of making a difference with each pixel in the transform window, a method using linearsegmentation to calculate the adaptive weight is used. The adaptive weight is also used in the stage of cost aggregation,the window size of that is determined by comparing the gradient value of the centerand itsleft and right two points. Theexperimental results show that the matching accuracy of the proposed algorithm is much better than the traditional Censusalgorithm, especially in the discontinuous depth region, the complexity of calculation and the difficulty of hardware implementationis not increased obviously.