摘要
针对运动恢复结构算法在摄像机拍摄过程中需要改变摄像机位置的问题,利用物体与摄像机的相对运动关系,获取包含物体多角度视图的图像序列。在图像序列中定位物体所在区域,约束特征点提取的适用范围,设计并实现一种相对运动恢复结构算法。实验结果表明,SFRM算法适用于摄像机固定位置的拍摄方式,能将位于物体区域的匹配对占总匹配对的比例从32%提高到84%,相较于传统的SFM算法在时间效率上得到38.29%的提升,成功扩展SFM算法的适用场景。
Aiming at the problem that the structure from motion algorithm needs to change the camera position in the process of camera photographing,the relative motion relationship between the target object and the camera is used to obtain image sequences containing multi angle views of target objects.A structure from relative motion algorithm is designed and implemented by locating the region of the target object in the image sequence and constraining the applicable range of feature point extraction.The experimental results show that SFRM algorithm is suitable for the photographing mode of fixed position camera.SFRM can increase the proportion of matching pairs located in the target object area from 32%to 84%of the total matching pairs.Compared with the traditional SFM algorithm,the time efficiency is improved by 38.29%.The available scenarios of SFRM algorithm is successfully expanded.
作者
高梓皓
张巧芬
王桂棠
温腾腾
庞亮雨
贾林锋
Gao Zihao;Zhang Qiaofen;Wang Guitang;Wen Tengteng;Pang Liangyu;Jia Linfeng(School of Electromechanical Engineering,Guangdong University of Technology,Guangzhou 510006,China;Cangke Intelligent Technology Co.,Ltd.,Foshan 528225,China)
出处
《电子测量技术》
北大核心
2023年第1期97-102,共6页
Electronic Measurement Technology
基金
国家自然科学基金(61705045)
佛山广工大研究院创新创业人才团队计划项目(20191108)资助
关键词
SFM算法
特征点提取
点云
三维重建
机器视觉
structure from motion
feature point detector
point cloud
3D reconstruction
machine vision
