一种新的改进Canny边缘检测算法

对Canny边缘检测算法的性能进行了分析和评价, 针对传统Canny算法在梯度幅值计算上的缺陷, 提出了一种通过计算像素八邻域内图像一阶偏导数有限差分来确定梯度幅值的方法, 改善了传统Canny算子利用2×2邻域内求有限差分均值对噪声的敏感性, 提高了边缘定位的精度. 在此基础上, 在基于非极值抑制原理检测边缘点时, 将像素的八邻域分为4个像限结合双线性插值的方法进一步提高了边缘检测的精度和准确度. 实验结果表明, 该算法在保证实时性的同时, 具有更好的检测精度和准确度.

An improved edge detection algorithm based on the Canny operator

The capability of edge detection using the Canny operator was improved. Gradient computation and the determination of edge points were improved. Pixel gradients were calculated using the first-order derivative within eight neighborhoods instead of four, and the location of the edge was determined via the non-maximal suppression process, in which four-quadrant bilinear interpolation was applied to the eight neighborhoods of pixel gradients, thus improving the precision of the detected edge. Experimental results show that the new algorithm is efficient for edge detection, especially for detection of the outer boundary of the iris.