摘要
立体匹配作为三维重建的关键步骤之一,针对基于Census变换的立体匹配算法在视差不连续区域以及弱纹理区域匹配精度较差,且易受光照不均与噪声干扰的问题,提出一种基于改进Census变换与分区域聚合的立体匹配算法。在Census变换阶段令窗口中所有像素点根据灰度值大小进行降序排序,选取灰度值为中位数的像素点替代窗口的中心像素;然后进行Hamming距离计算得到结果,再加权融合亮度或光照差的绝对值;接下来将代价聚合阶段的图像通过梯度大小划分为边缘和平滑区域,分别使用十字交叉域完成初始代价聚合;最终采用WTA找到初始视差值,经过一系列视差优化步骤获得最终视差图。在MiddleburyV3.0版本的测试平台上对该算法进行评测,实验结果表明,总体像素的视差平均误差为8.26%,相较于AD-Census算法下降了2.95%,具有较高的匹配精度,且对光照及噪声有较好的鲁棒性。
As one of the key steps in 3D reconstruction,stereo matching based on Census transform has poor matching accuracy in areas with discontinuous parallax and weak texture,and is prone to uneven illumination and noise interference.This paper proposes a stereo matching al-gorithm based on improved Census transform and regional aggregation.In the Census transform stage,all pixel points in the window are sorted in descending order according to the size of gray values,Select a pixel with a median grayscale value to replace the center pixel of the window.After that,the Hamming distance calculation is performed to obtain the result,and then the absolute value of the fusion brightness or illumina-tion difference is weighted.Then,the image in the cost aggregation stage is divided into edge and smooth regions by gradient size.The initial cost aggregation is completed using cross domain,respectively.Finally,the initial disparity value is found using WTA,and the final disparity map is obtained through a series of disparity optimization steps.The algorithm in this article is evaluated on the Middlebury V3.0 testing plat-form.The experimental results show that the average disparity error of the overall pixel is 8.26%,a decrease of 2.95%compared to the AD-Census algorithm,with high matching accuracy and good robustness to light and noise.
作者
胡鑫力
周锋
郭乃宏
姚凯文
李楠
王如刚
HU Xini;ZHOU Feng;GUO Naihong;YAO Kaiwen;LI Nan;WANG Rugang(School of Information Engineering,Yancheng Institute of Technology,Yancheng 224051,China;Yancheng Xiongying Precision Machinery Co.,LTD,Yancheng 224006,China)
出处
《软件导刊》
2024年第4期164-170,共7页
Software Guide
基金
国家自然科学基金项目(61673108)
江苏省自然科学基金项目(BK20181050)
江苏省产学研合作项目(BY2020335,BY2020358)
江苏省研究生实践创新项目(SJCX23_1873)
盐城工学院研究生实践创新计划项目(SJCX22-XZ037,SJCX22_XY059)。
