基于图像噪声检测的平滑方法研究  

Research on Smoothing Method Based on Image Noise Detection

作　　者：娄联堂[1] 吴高林[1] 

机构地区：[1]中南民族大学数学与统计学学院,武汉430074

出　　处：《中南民族大学学报（自然科学版）》2017年第4期137-142,共6页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：国家自然科学基金资助项目(60975011);中央高校基本科研业务费专项资金资助项目(CZW15051;YZZ13003);中南民族大学研究生科研创新基金资助项目(2016sycxjj137)

摘　　要：针对含有椒盐噪声的图像,首先利用了TSMD阈值方法将图像二值化,有效地检测了图像的噪声,然后估算出了噪声图像的SNR,比较了与原噪声图像SNR的差异程度,得到了相应地质量评价方法,最后通过选择性中值滤波的方法去除了图像上的噪声.结果表明:相比于全局中值滤波方法,该方法能够有效地去除椒盐噪声,保留更多的图像细节,对图像的影响较小.In this paper,firstly,the TSMD threshold method was used to binarize the image to effectively detect the noise of the image containing salt and pepper noise.Then the SNR of the noise image was estimated,and the difference with the original noise image was compared,the corresponding quality evaluation method was obtained.Finally,the noise on the image was removed by the selective median filter method.The results showed that,compared to the global median filter method,this method can effectively remove the salt and pepper noise,retain more details of the image and has less impact on the image.

关 键 词：噪声检测 图像平滑 TSMD阈值方法 质量评价 选择性中值滤波 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]