中文摘要：

鱼眼镜头拍摄的图像具有严重的变形，要想利用鱼眼图像的信息，就必须将其校正为常规的线性图像。提取鱼眼图像有效区域是进行校正的前提，同时，鱼眼图像有效区域提取的精度将影响到鱼眼图像校正的效果。传统最常用的方法是扫描线逼近法，计算量较小，速度较快，但也具有抗干扰性差的缺陷。针对有效区域边界存在黑色像素点影响传统算法精度的缺点，提出通过改变扫描步长，垂线缩进的改进算法。实验结果表明，改进算法能准确确定有效区域的中心坐标和半径，具有精度高、耗时少、满足实时性要求的特点，具有较强的实用价值。

英文摘要：

Since the image taken by the fish-eye lens has serious distortion, in order to take advantage of the information of the fish-eye image, it must be corrected as the conventional linear image. The extraction of the fish-eye effective area is a prerequisite for correction. The extraction accuracy of the fish-eye image effective area can affect on the correction effect of the fish-eye image. The scanning line approximation method of the commonly-used ones has small calculated amount and fast speed, as well as poor anti-interference ability. To eliminate the shortcoming that the black pixel existing in the effective area boundary can affect on the accuracy of the traditional algorithms, an improved algorithm of changing the scanning step size and retracting the vertical line is proposed. The experimental results show that the algorithm can determine the center coordinate and radius of the active area accurately, has high precision and short time-consuming, can meet the real-time requirements, and has strong practical value.