Fast stereo matching using rectangular subregioning and 3D maximum-surface techniques

Fast stereo matching using rectangular subregioning and 3D maximum-surface techniques
复制标题

DOI:
10.1023/a:1014585622703
复制
发表时间:
2002-04-01
影响因子:
19.5
通讯作者:
Sun, CM
Sun, CM
中科院分区:
计算机科学2区
文献类型:
--
作者:
Sun, CM

文献摘要

被引文献

相似文献

本文提出了一种快速可靠的立体匹配算法,该算法采用快速互相关,矩形子区域(RSR)和三维最大表面技术,在一个由粗到精的计划产生一个密集的视差图。快速相关是通过使用盒滤波技术,其速度是不变的相关窗口的大小,并通过分割立体图像成矩形子图像在不同层次的金字塔。通过使用矩形子图像,不仅可以进一步提高相关的速度,还可以减少中间存储器存储需求。通过获得全局3D最大表面而不是简单地选择给出每个像素的局部最大相关系数值的位置,在3D相关系数体积中找到立体图像的视差图。使用我们新的两阶段动态规划(TSDP)技术获得3D最大表面。本文有两个原创性的贡献:(1)RSR技术的快速相似性度量的发展;(2)TSDP技术的发展,有效地获得三维最大表面在一个三维体积。我们的算法在512 x 512图像上用C语言实现的典型运行时间在500 MHz PC上大约为几秒钟。对多种合成和真实的图像进行了测试,取得了良好的效果。
This paper presents a fast and reliable stereo matching algorithm which produces a dense disparity map by using fast cross correlation, rectangular subregioning (RSR) and 3D maximum-surface techniques in a coarse-to-fine scheme. Fast correlation is achieved by using the box-filtering technique whose speed is invariant to the size of the correlation window and by segmenting the stereo images into rectangular subimages at different levels of the pyramid. By working with rectangular subimages, not only can the speed of the correlation be further increased, the intermediate memory storage requirement can also be reduced. The disparity map for the stereo images is found in the 3D correlation coefficient volume by obtaining the global 3D maximum-surface rather than simply choosing the position that gives the local maximum correlation coefficient value for each pixel. The 3D maximum-surface is obtained using our new two-stage dynamic programming (TSDP) technique. There are two original contributions in this paper: (1) development of the RSR technique for fast similarity measure; and (2) development of the TSDP technique for efficiently obtaining 3D maximum-surface in a 3D volume. Typical running time of our algorithm implemented in the C language on a 512 x 512 image is in the order of a few seconds on a 500 MHz PC. A variety of synthetic and real images have been tested, and good results have been obtained.