An Improved Method for Feature Point Matching in 3D Reconstruction

被引：2

作者：

Wang, Zhongren ^{[1
,2
]}

Quan, Yanming

机构：

[1] South China Univ Technol, Sch Mech & Automot Engn, Guangzhou 510641, Guangdong, Peoples R China

[2] Xiangfan Univ, Dept Mech Engn, Xiangfan AH-441053, Peoples R China

来源：

ISISE 2008: INTERNATIONAL SYMPOSIUM ON INFORMATION SCIENCE AND ENGINEERING, VOL 1 | 2008年

关键词：

Feature point matching; fundamental matrix; homography matrix; MAPSACLM; 3D reconstruction;

D O I：

10.1109/ISISE.2008.239

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, a MAPSACLM algorithm for feature point matching is proposed. This method integrates the MAPSAC algorithm with nonlinear optimization by using the results of MAPSAC as the initial value of the fundamental matrix and homography matrix. Firstly, gray level cross-correlation matching method was used to realize initial matching. Secondly, the fundamental matrix and the homography matrix were estimated robustly with MAPSAC algorithm. As a result, most of the outliers were detected and removed. Then, nonlinearly optimized fundamental matrix and homography matrix by Levenberg-Marquardt algorithm were used to obtain more precise matching points. Lots of experiments show that this algorithm is efficient and it improves the robustness and accuracy of the automatic image matching in 3D reconstruction.

引用

页码：159 / +

页数：2

共 8 条

[1] A pixel dissimilarity measure that is insensitive to image sampling [J].