【摘要】 基于视频信号处理中用于场内插值的边缘自适应平均算法(ELA)思想,提出一种针对椒盐噪声图像滤波的改进型边缘自适应平均去噪算法(IELAD)。算法在扩展横向5+5阶ELA窗口的基础上增加了纵向5+5阶ELA窗口。根据噪声点所处边缘方向,自适应地确定窗口及滤波方向进行滤波处理。实验表明IELAD算法充分利用了ELA窗口的保边特性,能够在去除椒盐噪声的同时有效地保持图像的边缘及细节信息。更多还原

【Abstract】 An ideal image denoising algorithm should preserve details of image while removing the image noises.This article drew on the Edge-based Line Average algorithm(ELA)thinking which is used to venue interpolate in video signal processing and put forward an Improved Edge-based Line Average Denoising algorithm(IELAD)which is used to remove salt-and-pepper noises.In this algorithm,an extended 5+5order vertical-window is added based on the 5+5order horizontal-window.We saw the noise pixels as the interpolating pixels.Then according to the edge direction of the noise,the right window and direction were choosed adaptively for filtering.Experiments show that this method makes full use of the edge-preserving property of ELA window and can effectively remove the salt-and-pepper noises.This algorithm can maintain the edge details of the image and reduce image blur.更多还原