一种新的强抗噪边缘检测方法  

A NEW EDGE DETECTION METHOD WITH HIGH ANTI-NOISE PERFORMANCE

作　　者：雷涛[1] 王小鹏[1] 王静[1] 

机构地区：[1]兰州交通大学电子与信息工程学院,甘肃兰州730070

出　　处：《计算机应用与软件》2009年第3期65-67,共3页Computer Applications and Software

摘　　要：针对传统形态学边缘检测方法存在去除噪声与边缘细节保留矛盾的问题,提出了一种能有效去除噪声且准确检测图像边缘的方法。该方法首先利用多结构元素对图像进行滤波,然后用全方位多角度线结构元素结合一种改进的形态学算子检测图像边缘,最后将所得结果进行自适应权重求和,得到噪声存在下较理想的图像边缘。实验表明,该方法对噪声具有较强的抑制能力,在准确检测图像边缘、保留更多细节边缘的同时,能够有效地去除图像中的多种噪声,具有较高的信噪比。To solve the difficulty of edge sharpness preserving in contradiction to noise removing in traditional morphological edge detection, a new edge detection method is proposed to remove noise efficiently as well as to detect image edge more accurately. With the new method, the input image is filtered by multi-structure element; its edge is then detected by structure element from all directions and multi-angle in combination with an improved morphological operator, and the ideal image edge is finally obtained under the noisy condition through summation of the detected edge results with adaptive weight. Experiments show that this method has better ability in restraining image noise, it can efficiently remove various noises with finer edges sharpness preserved when accurately detecting the image edge, and improves SNR as well.

关 键 词：边缘检测 抗噪声 形态学滤波 自适应权重 信噪比 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP391[自动化与计算机技术—计算机科学与技术]