基于梯度的自适应边缘检测算法研究  

Research on self-adaptive algorithm of edge detection based on gradient

作　　者：李海华[1] 范娟[1] 

机构地区：[1]华中科技大学文华学院机电学部,武汉430074

出　　处：《河南理工大学学报（自然科学版）》2013年第1期76-79,共4页Journal of Henan Polytechnic University（Natural Science）

摘　　要：对常见的边缘检测算法进行改进,基于梯度提出了一种应用于灰度图像的自适应阈值边缘检测算法.根据边界点像素灰度值的差异,分析模版中的9个像素,计算出像素的梯度幅度和梯度方向;按梯度值的不同将图像分割成若干个区域,计算每个区域的灰度平均值,确定阈值,实现边缘检测.实验表明,该方法检测出的边缘更细、更准确,可以除去虚假的边缘,是一种有效的对灰度图像进行边缘检测的方法.To improve a common edge detection algorithm, a new method of self-adaptive threshold applying to gray level images is present based on the gradient. According to the different gray levels of the pixels on the edge, the gradient magnitude and direction of each pixel can be calculated after analyzing the nine pixels in the mask. The image is divided into several regions by different gradient values. The edge detection can be implemented when we figure out the average gray level for each region and determine the threshold. The exper- iment shows that this method can detect thinner and more accurate edges, and it can also eliminate the false edges. It is an effective method to detect the edge for a gray level image.

关 键 词：梯度 自适应阈值 边缘检测 图像处理 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]