摘要
无人机全景摄像机在飞行环境中,多角度拍摄进行图像全景拼接时,存在视差导致的拼接误差大等问题。为此本文对无人机全景摄像机关键技术展开了研究,提出一种基于相机标定的柱面图像拼接技术,这种技术可用于全景拍摄系统,大大提高了系统的实时性,同时也保证了拼接的准确性。首先是对拍摄所采用的广角镜头进行标定,再对采集的图像进行畸变矫正,重投影获取柱面序列图像,最后基于块匹配算法进行图像配准和线性融合,得到高质量的拼接图像。
UAV panoramic camera in the flight environment, multi-angle shooting image panoramic stitching, there is parallax stitching error caused by large problems. In this paper, the key technology of UAV panoramic camera is studied, and a cylindrical image stitching technology based on camera calibration is proposed. This technology can be used in panoramic shooting system, which greatly improves the real-time performance of the system and ensures the accuracy of stitching. Firstly, the wide-angle lens used for shooting is calibrated. Then, the collected image is corrected for distortion, and the cylindrical sequence image is obtained by reprojection. Finally, the image registration and linear fusion are carried out based on the block matching algorithm to obtain the high-quality mosaic image.
作者
葛思远
Ge Siyuan(College of Electronic Information Engineering,Shenyang Aerospace University,Shenyang 110136)
出处
《现代计算机》
2021年第36期39-43,共5页
Modern Computer
关键词
全景摄像机
相机标定
畸变矫正
图像配准
图像融合
panoramic camera
camera calibration
distortion correction
image registration
image fusion
