摘要
针对图像椒盐噪声检测中的不确定性问题,利用基于证据理论的椒盐噪声检测算法,并联合Holt双参数指数平滑法对噪声图像进行滤除.依据椒盐噪声的极端性和不连续性,在检测噪声过程中利用证据理论的不确定性判据给出两种准则,并最终给出决策融合方法来判别噪声.此外,根据预检测结果给出噪声点的消除方法,实现图像消噪功能.仿真实验验证了本文方法的有效性.
The detection of salt and pepper noise was an uncertainty problem in the image processing.In this paper,the evidence theory was used to detect salt and pepper noise and Holt′s two-parameter exponential smoothing method was used to filter out noise of images.According to the extremeness and discontinuity of salt and pepper noise,two uncertainty criterions based evidence theory were given and a decision fusion method was given to distinguish noise in the process of detecting noise.In addition,a method for eliminating noise points based on the pre-detection results was given to achieve image denoising.Simulation experiments verified the effectiveness of this method.
作者
王金凤
WANG Jin-feng(School of Science,Northeast Forestry University,Harbin 150040,China)
出处
《哈尔滨商业大学学报（自然科学版）》
CAS
2021年第6期689-694,共6页
Journal of Harbin University of Commerce：Natural Sciences Edition
关键词
图像去噪
椒盐噪声
证据理论
双参数指数平滑
滤波算法
决策融合
image denoising
salt and pepper noise
evidence theory
two-parameter exponential smoothing
filtering algorithms
decision fusion
