In this paper,for three-linear array stereo-imagery of Mapping Satellite-1,semi-global matching algorithm based on mutual information is researched and used in three-dimensional reconstruction.The pyramid and multiple-baseline matching strategy is adopted and mutual informationof per-pixel is regarded as matching unit and full matching is realized,according to the principle of energy function minimization.The experimental results show that the proposed method can accurately realize three-dimensional reconstruction for Mapping Satellite-1 imagery.
WANG Yuqin
,
LI Heyuan
. Three-dimensional Reconstruction of Mapping Satellite-1 Imagery Using Semi-global Matching Algorithm Based on Mutual Information[J]. Forest and Grassland Resources Research, 2016
, 0(4)
: 134
-139
.
DOI: 10.13466/j.cnki.lyzygl.2016.04.024
[1] 高波.从双目立体图像中恢复三维信息的研究[D].上海:上海交通大学,2007.
[2] Hirschmuller H. Stereo Processing by Semi-global Matching and Mutual Information[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(2):328-341.
[3] Birchield S,Tomasi C.Depth discontinuities by pixel-to-pixel stereo[J].International Conference on Computer Vision,1998,35(3):1073-1080.
[4] 张力,张继贤.基于多基线影像匹配的高分辨率遥感影像DEM的自动生成[J].武汉大学学报:信息科学版,2008,3(9):943-946.
[5] 纪松,范大昭,戴晨光,等.线阵影像GC3 多视匹配及其扩展模型研究[J].测绘科学技术学报,2009,26(6):44-47.
[6] 范大昭.多线阵影像匹配生成DSM的理论与算法[D].郑州:信息工程大学测绘学院,2007.
[7] 纪松.多视匹配策略与优化方法研究[D].郑州:信息工程大学测绘学院,2012.
