摘要
对数字图像进行去噪操作是图像预处理过程中的重要组成部分。研究了在使用不同的非线性滤波器对图像进行去噪操作后对于图像边缘信息的保留效果。实验选取中值滤波、BM3D滤波器和双边滤波器3种非线性滤波器对被高斯噪声和椒盐噪声所污染的图片进行去噪操作。实验结果表明,在只有高斯噪声的情况下,3种滤波器都能收获较好的滤波效果,而在只有椒盐噪声的情况下,双边滤波的去噪效果较差。在3组实验中,BM3D在对图片进行去噪时对于边缘信息的保持最好,而中值滤波和双边滤波会损失较多的图片边缘信息,在对图片进行滤波时,BM3D消耗的时间资源较多。
Denoising the image of the drone is an important part of the digital image preprocessing process.The amount of image information retained after denoising can directly affect the effect of subsequent image processing.The edge information of the image is the image information.An important part of the ring.Therefore,in order to ensure that more image edge information can be preserved when denoising the image,it is of great significance to study the image edge information after denoising the image using different nonlinear filters.The purpose of the experiment is to study the denoising effect of median filtering,BM3 D filter and bilateral filter and the preservation effect of image edge information in the case of selecting different noises,and record the calculation time of the two algorithms.The experimental results show that under the condition of only Gaussian noise,all three filters can obtain better filtering effect,and when there is salt and pepper noise,the denoising effect of bilateral filtering is poor.In the three sets of experiments,BM3 D keeps the best edge information when denoising the picture,while median filtering and bilateral filtering will lose more picture edge information.When filtering the picture,the time resource consumed by BM3 D More.
作者
仇俊杰
李大军
QIU Junjie;LI Dajun(School of Surveying and Mapping Enginerring,East China University of Technology,330013,Nanchang,PRC)
出处
《江西科学》
2020年第2期269-274,共6页
Jiangxi Science
关键词
非线性滤波
中值滤波
BM3D
双边滤波
边缘保持
nonlinear filtering
median filtering
BM3D
bilateral filtering
edge retention