Stereo Matching Based on Dissimilar Intensity Support and Belief Propagation

被引：3
|
作者：
Da, Feipeng [1 ]
He, Fu [1 ]
Chen, Zhangwen [1 ]
机构：
[1] Southeast Univ, Key Lab Measurement & Control Complex Syst, Minist Educ, Sch Automat, Nanjing 210096, Jiangsu, Peoples R China
来源：
JOURNAL OF MATHEMATICAL IMAGING AND VISION | 2013年 / 47卷 / 1-2期
关键词：
Stereo matching; Local edge detection; Dissimilar intensity support; Belief propagation optimization; Cost aggregation; ALGORITHM; WEIGHT;
D O I：
10.1007/s10851-013-0448-1
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
A novel algorithm based on the window construction method using local edge detection is presented. Firstly, in order to construct the adaptive window, a virtual closed edge is formed around each pixel via second order differential operator. Secondly, a novel rule called Dissimilar Intensity Support (DIS) technique is proposed. This rule is used to distinguish support pixels with dissimilar intensity from those with similar intensity for each centered pixel. So that the performance of window-based cost aggregation computation is improved. Thirdly, belief propagation (BP) optimization algorithm is used to obtain the disparity. The experimental results based on Middlebury stereo benchmark show that the proposed algorithm has good performances.
引用
收藏
页码：27 / 34
页数：8
相关论文
共 50 条
  • [41] Stereo Matching with Color-Weighted Correlation, Hierarchical Belief Propagation, and Occlusion Handling
    Yang, Qingxiong
    Wang, Liang
    Yang, Ruigang
    Stewenius, Henrik
    Nister, David
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2009, 31 (03) : 492 - 504
  • [42] Stereo Matching by Filtering-Based Disparity Propagation
    Wang, Xingzheng
    Tian, Yushi
    Wang, Haoqian
    Zhang, Yongbing
    PLOS ONE, 2016, 11 (09):
  • [43] Intensity and feature based stereo matching by disparity parameterization
    Wei, GQ
    Hirzinger, G
    SIXTH INTERNATIONAL CONFERENCE ON COMPUTER VISION, 1998, : 1035 - 1040
  • [44] A novel method for graph matching based on belief propagation
    Lin, Xue
    Niu, Dongmei
    Zhao, Xiuyang
    Yang, Bo
    Zhang, Caiming
    NEUROCOMPUTING, 2019, 325 : 131 - 141
  • [45] Enhancement of sparse silicon retina-based stereo matching using belief propagation and two-stage postfiltering
    Kogler, Juergen
    Eibensteiner, Florian
    Humenberger, Martin
    Sulzbachner, Christoph
    Gelautz, Margrit
    Scharinger, Josef
    JOURNAL OF ELECTRONIC IMAGING, 2014, 23 (04)
  • [46] A 30fps Stereo Matching Processor Based on Belief Propagation with Disparity-Parallel PE Array Architecture
    Park, Junyoung
    Lee, Seungjin
    Yoo, Hoi-Jun
    2010 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, 2010, : 453 - 456
  • [47] Event-Based Stereo Depth Estimation Using Belief Propagation
    Xie, Zhen
    Chen, Shengyong
    Orchard, Garrick
    FRONTIERS IN NEUROSCIENCE, 2017, 11
  • [48] Stereo Line Matching Based on the Combination of Geometric and Intensity Data
    Woo, Dong-Min
    Park, Dong-Chul
    2009 24TH INTERNATIONAL SYMPOSIUM ON COMPUTER AND INFORMATION SCIENCES, 2009, : 579 - 583
  • [49] Improved SVD matching based on Delaunay triangulation and belief propagation
    Dou, Jianfang
    Qin, Qin
    Tu, Zimei
    OPTIK, 2015, 126 (24): : 4861 - 4867
  • [50] Area matching based on belief propagation with applications to face modeling
    Onofrio, D
    Sarti, A
    Tubaro, SW
    ICIP: 2004 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1- 5, 2004, : 1943 - 1946
12345