基于KLT多视图特征跟踪的物体三维重建  

Object three-dimensional reconstruction based on KLT multi-view feature tracking

作　　者：吴正斌 彭业萍 曹广忠[1,2] 曹树鹏 WU Zhengbin;PENG Yeping;CAO Guangzhong;CAO Shupeng(College of Mechatronics and Control Engineering, Shenzhen University, Shenzhen 518060, China;Guangdong Key Laboratory of Electromagnetic Control and Intelligent Robot, Shenzhen University, Shenzhen 518060, China)

机构地区：[1]深圳大学机电与控制工程学院,广东深圳518060 [2]深圳大学广东省电磁控制与智能机器人重点实验室,广东深圳518060

出　　处：《合肥工业大学学报（自然科学版）》2020年第7期901-906,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金青年科学基金资助项目(51905351,U1813212);广东省自然科学基金资助项目(2018A030310522);深圳市科技计划资助项目(JCYJ20190808113413430,JCYJ20170818100522101)。

摘　　要：为了提高三维重建精度,文章提出一种基于Kanade-Lucas-Tomasi(KLT)多视图特征跟踪的物体三维重建方法。首先采用Shi-Tomasi算法提取特征,通过KLT跟踪算法获取特征匹配点,从而将所有相邻图像进行三维点云重建,并通过光束法平差(bundle adjustment,BA)算法进行优化;然后采用KLT跟踪算法获取空间点在相邻3幅图像的投影点,从而建立不同视图坐标系间的变换模型,获取转换比例系数;再通过融合多视图的三维点云实现物体的三维重建。实验结果表明,通过KLT算法,不仅提高了特征点匹配数量,还解决了多视图构建的不同三维点云坐标系的相互转换问题,实现了物体的多视图三维重建。In order to improve the accuracy of three-dimensional reconstruction,this paper proposes a three-dimensional reconstruction method based on Kanade-Lucas-Tomasi(KLT)multi-view feature tracking.Firstly,Shi-Tomasi algorithm is used to extract features,and KLT tracking algorithm is used to obtain feature matching points,so as to reconstruct all adjacent images in three-dimensional point cloud,and then the bundle adjustment(BA)algorithm is used to optimize the reconstruction results.Secondly,KLT tracking algorithm is adopted to obtain the projection points of spatial points in three adjacent images,so as to establish the transformation model between coordinate systems of different views,and obtain the conversion proportional coefficient.Finally,the three-dimensional reconstruction of the object is realized by integrating the multi-view three-dimensional point cloud.The experimental results show that the KLT algorithm can not only improve the number of feature point matching,but also solve the problem of the transformation between different three-dimensional point cloud coordinate systems constructed by multiple views and realize the three-dimensional reconstruction of multi-view objects.

关 键 词：三维重建 多视图 特征提取 Kanade-Lucas-Tomasi(KLT)跟踪 点云融合 

分 类 号：TP391.414[自动化与计算机技术—计算机应用技术]