Direction-Based Adaptive Switching Filter for Removing High-Density Impulse Noise  

作　　者：刘会刚 孙菁 张福海 任立儒 

机构地区：[1]Tianjin Key Laboratory for Photo-Electronic Thin Film Devices and Technology, College of Electronic Information and Optical Engineering, Nankai University

出　　处：《Transactions of Tianjin University》2014年第6期422-428,共7页天津大学学报（英文版）

基　　金：Supported by the National Natural Science Foundation of China(No.61401237);the Natural Science Foundation of Tianjin(No.13JCQNJC01200);the Specialized Research Fund for the Doctoral Program of Higher Education of China(No.20130031120034)

摘　　要：A direction-based adaptive switching(DBAS) filter is presented for the removal of high-density impulse noise in images. The extrema detection and 28-directional detection are employed to discriminate the pixels as noisy or noise-free. If a pixel is classified as noisy, it will be replaced by a median or a mean value within an adaptive filter window with respect to different noise densities. Simulation results show that the miss-detection ratio and false-alarm ratio are both very low even at noise level as high as 90%. At the same time, better results are obtained in terms of the qualitative and quantitative measures. The peak signal-to-noise ratios increase by nearly 1 dB compared with other existing algorithms. In addition, the computation time is around 10 s for test images with resolutions of 512×512since the proposed approach has low complexity.A direction-based adaptive switching(DBAS) filter is presented for the removal of high-density impulse noise in images. The extrema detection and 28-directional detection are employed to discriminate the pixels as noisy or noise-free. If a pixel is classified as noisy, it will be replaced by a median or a mean value within an adaptive filter window with respect to different noise densities. Simulation results show that the miss-detection ratio and false-alarm ratio are both very low even at noise level as high as 90%. At the same time, better results are obtained in terms of the qualitative and quantitative measures. The peak signal-to-noise ratios increase by nearly 1 dB compared with other existing algorithms. In addition, the computation time is around 10 s for test images with resolutions of 512×512since the proposed approach has low complexity.

关 键 词：direction-based FILTER impulse noise noise detection nonlinear FILTER MEDIAN FILTER 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TN713[自动化与计算机技术—计算机科学与技术]