摘要： 以图像二值化为前提，使用基于区域生长的改进算法，获取图像轮廓，实现鱼眼图像校正，克服鱼眼图像轮廓区域内出现大量黑色像素使面积统计算法失效的缺点。仿真实验表明，该算法精度高、效率高，能够获得满意的校正结果。

关键词: 鱼眼图像, 区域生长, 轮廓提取, 非线性最小二乘拟合

Abstract: On the premise of image binary, image contour is obtained by means of an improved algorithm based on region growing, image correction can be completed. It can overcome the shortage that massive black pixels appear in the fisheye image circular outline area and make the statistical algorithm fail. Simulation results show that the proposed algorithm has advantage of high precision and high efficiency, satisfactory results are obtained.

Key words: fisheye image, region growing, contour extraction, nonlinear least square fitting

中图分类号: 

• TP391.41