摘要
立体匹配是三维重建中重要的组成部分。针对DP匹配算法生成视差图时效性和准确性偏低的问题,采用SGBM算法获取视差图,并进行三维重建。固定两个单目相机于铝板上,通过棋盘格进行标定,得到相机内外参与畸变系数,并利用标定结果进行立体校正。增强双目相机获取到的场景照片质量,降低外界因素的影响。获取视差图,并结合内参数通过三角变换原理获取空间点云。通过实验对比表明,SGBM算法生成的视差图误匹配率低,时效性高。
Stereo matching is an integral part of the 3 D reconstruction.In view of the low timeliness and accuracy of the disparity map generated by the DP matching algorithm,this paper uses the SGBM algorithm to obtain the disparity map and carry out 3 D reconstruction.Two monocular cameras are fixed on the aluminium plate and calibrated by checkerboard.The distortion coefficients of the camera inside and outside are obtained,and the calibration results are used for stereo correction.Enhance the quality of the scene photos obtained by the binocular camera and reduce external factors’influence.The parallax map is obtained,and the spatial point cloud is obtained by trigonometric transformation combined with internal parameters.The experimental results show that the disparity map generated by the SGMB algorithm has a low mismatching rate and high timeliness.
作者
赵成星
张晓玲
杨宇
ZHAO Chengxing;ZHANG Xiaoling;YANG Yu(School of Electrical Information Engineering,Jiangsu University of Technology,Changzhou Jiangsu 213000,China)
出处
《激光杂志》
CAS
北大核心
2021年第4期139-143,共5页
Laser Journal
基金
国家自然科学基金(No.61305123)。
关键词
相机标定
立体匹配
SGBM算法
三维重建
camera calibration
stereo matching
SGBM algorithm
3D reconstruction
