一种改进的Canny边缘检测AGT算法  

AN IMPROVED CANNY EDGE DETECTION AGT ALGORITHM

作　　者：冯珂[1] 朱敏[1] 钟煜[1] 范量[1] 

机构地区：[1]四川大学计算机学院,四川成都610064

出　　处：《计算机应用与软件》2012年第3期265-266,300,共3页Computer Applications and Software

摘　　要：使用Canny算子进行图像边缘检测。由于阈值参数手动输入,边缘检测效果受到影响,为此,提出一种自动生成阈值的算法:AGT(automatically generate threshold)算法。该算法根据图像自身灰度均值和方差均值信息,可实现Canny算子的高低阈值自动计算,使Canny算子不需输入阈值直接得到精确边缘。实验结果显示,该算法不仅能得到较好的边缘检测效果,并且有很强的自适应性。When we use Canny operator to detect the edge of the image,the effect may be impacted since the threshold of parameters are inputted manually.This paper therefore proposes an AGT（automatically generate threshold） algorithm.According to the information of mean gray degree and mean variance of the image itself,the algorithm can automatically calculate the high and low threshold of Canny operators,which enables the Canny operators directly obtaining accurate edge without inputting the threshold.The result of experiments shows that this algorithm can get a good edge detection effect and also has strong adaptability.

关 键 词：边缘检测 CANNY算子 自适应 阈值 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]