激光与光电子学进展, 2016, 53 (9): 091501, 网络出版: 2016-09-14   

基于迭代校正的鱼眼棋盘格图像角点检测 下载: 1229次

Corner Detection for Fisheye Checkerboard Images Based on Iterative Correction
作者单位
1 上海大学机电工程与自动化学院, 上海 200072
2 中国科学院宁波材料技术与工程研究所, 浙江 宁波 315201
摘要
传统的棋盘格角点检测方法难以准确提取鱼眼图像边缘处的角点,造成相机标定精度低,为此提出一种面向鱼眼镜头大畸变成像的棋盘格角点检测方法。该方法通过多次拍摄和逐级迭代估计和优化相机参数,利用远距中心小畸变区域的图像获取相机初始参数,通过空间坐标变换和像素点灰度插值实现近距大畸变图像畸变补偿,进而实现对图像边缘处各角点的检测,并根据参数映射关系计算出近距大畸变图中所对应的角点坐标。仿真实验与真实鱼眼图像实验结果表明,该方法实现简单,能够有效提高角点检测的数量和质量,满足实际应用需求。
Abstract
There is a challenging problem during camera calibration with fisheye lens using the traditional checkerboard corner detection method, and leads to low calibration precision of camera. So, an iterative correction checkerboard corner detection method for large distortion images is proposed. The main idea of this method is step by step iterative estimation and optimization of camera parameters through several shots. The camera initial parameters are obtained using the images from the remote-center small-distortion region. The distortion compensation of near-distance large distortion is realized through space and coordinate transformation and pixel interpolation gray. On this basis, the corner points of image edges can be detected, and the corner coordinates corresponding to the near-distance large-distortion images are calculated based on the parameter mapping relationship. Simulation and real image experimental results show that the proposed method is simple to be implemented, and improves the quantity and quality of corner detection effectively, which can satisfy the practical application requirements.

申夏晶, 程梦娇, 肖江剑, 宋文祥. 基于迭代校正的鱼眼棋盘格图像角点检测[J]. 激光与光电子学进展, 2016, 53(9): 091501. Shen Xiajing, Cheng Mengjiao, Xiao Jiangjian, Song Wenxiang. Corner Detection for Fisheye Checkerboard Images Based on Iterative Correction[J]. Laser & Optoelectronics Progress, 2016, 53(9): 091501.

本文已被 5 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!