分块平均的噪声图像边缘检测

边缘检测是图像处理中很重要的组成部分,其效果好坏直接影响到后面的图像分析。针对传统的边缘检测算子对噪声的干扰较大,本文提出一种被噪声污染的图像边缘检测的新方法。将模板窗口平均分成不同方向的8块,计算不同方向的平均灰度值的绝对差,然后判断窗口中心像素是否为边缘点。实验结果表明,该边缘检测算子在抑制噪声对图像边缘的影响和保持图像细节方面,优于传统的边缘检测算子。

Average block of noise image edge detection

Image edge detection is a key component of image processing systems which strongly affects the iamge analysis.Traditional edge detector interfere with noise.This paper proposed a new edge detection method for noised image.The template will be devided into eight blocks with different directions.Then determined whether the center pixel of the window is the edge point throngh average gray value.The experiment results compared with traditional methods shows that this method remains good edge of image and can im...