摘要
本文提出了一个标定鱼眼相机参数的框架,通过利用对已知空间位置的标定图案拍摄所得到的图像信息,迭代优化相机的刚性变化(旋转矩阵及平移)以及鱼眼投影模型中的参数.使用迭代最近点(ICP)的方法求出相机对于世界坐标系的刚性变换即旋转矩阵及平移量,利用l_1回归方法优化鱼眼投影模型的相关参数.通过模拟数据及真实数据的实验,本文的算法所得结果可以直接用于校正后图像的拼接.
In this paper, we prepare a framework to estimate the fisheye camera calibration from the dot pattern. In this framework, we optimize the rigid transformation and the coefficient of the fisheye projection iteratively. The rotation and the translation of the fisheye camera are optimized by the iterative closest point（ICP） method, and the l1-regression is used to optimize the parameters of the fisheye projection. Out method is suit for all the normal fisheye projection. Both computer simulation and real data are used to test the proposed technique and very good results have been obtained. The results of our method can be used to solve the image mosaic problem as well.
作者
敖乃翔
李浩
王德勇
AO Nai-xiang LI Hao WANG De-yong(China Academy of Electronics and Information Technology, Beijing 100041 School of Mathematical Sciences, University of Science and Technology of China, Hefei 230026)
出处
《中国电子科学研究院学报》
北大核心
2017年第1期67-72,110,共7页
Journal of China Academy of Electronics and Information Technology
基金
科技部国家重点研发计划"人员多维特征采集数据门研制"
项目编号:2016YFC0800503
关键词
鱼眼图像
参数标定
迭代最近点
l1回归
fisheye
camera calibration
iterative closest point
l1-regression
