摘要
鉴于传统Canny边缘检测算法在高斯滤波方差和高低阈值选取上需要人工干预,不具备自适应能力。文章提出了一种通过迭代分割求取最佳高、低阈值的方法,改善了人为设定阈值自适应性不强的缺点,提高了边缘定位的精度。与此同时,还将检测结果与原Canny、Sobel和Log等边缘检测算子对图像的处理结果进行了比较,实验结果表明,采用改进的Canny边缘检测算法可以得到较为完整、清晰的边缘轮廓,具有更好的检测精度和准确度。
Because the variance of Gaussian filter and high and low thresholds should be determined artificially,Canny algorithm has no adaptive capacity. This paper proposes a method to obtain the optimal high and low thresholds using iterative segmentation. This method improved edge location precision as well as the weakness of setting threshold artificially.At the same time,We also compared the detection results with the traditional Canny, Sobel and Log edge detection operator for image processing of results. The experimental results show that the improved Canny can obtain more complete and clearer edge profile and has a better detection precision and accuracy.
出处
《电脑与信息技术》
2015年第2期4-7,21,共5页
Computer and Information Technology
