摘要
为了克服传统小波阈值函数存在固定偏差和不连续的问题,提出了一种改进的小波阈值函数,将改进的小波阈值方法和软阈值、硬阈值、Birge-Massart策略软阈值、半软阈值、半软阈值组合均值滤波等方法比较,峰值信噪比提高了1.2%~6.9%,均方根误差减小了4.8%~22.4%.在混有高斯噪声、椒盐噪声和斑点噪声的图像中,改进的小波阈值函数和自适应中值滤波组合的方法较单一的自适应中值滤波,峰值信噪比提高3 dB,均方根误差减小27%.结果表明,该算法有效的提升了高斯噪声图像的去噪效果,而且组合去噪方法对图像中的混合噪声去除效果较好.
In order to overcome the problems of fixed deviation and discontinuity of the traditional wavelet threshold function,an improved wavelet threshold function is proposed.Compared with soft threshold method,hard threshold method,Birge MassArt strategy soft threshold method,half soft threshold method and half soft threshold combined mean filtering method,the peak signal-to-noise ratio is increased by 1.2%~6.9%,and the root mean square error is reduced by 4.8%~22.4%.In the image mixed with Gaussian noise,salt and pepper noise and speckle noise,the combination method of improved wavelet threshold function and adaptive median filter is better than the single adaptive median filter,the peak signal-to-noise ratio is increased by 3 dB,and the root mean square error is reduced by 27%.The results show that the algorithm effectively improves the denoising effect of Gaussian noise image,and the combined denoising method has a better effect on the removal of mixed noise in the image.
作者
贾文良
陈雨
陈强
JIA Wen-liang;CHEN Yu;CHEN Qiang(Intelligent Control Institute,College of Electronics and Information,Sichuan University,Chengdu 610065,China)
出处
《微电子学与计算机》
北大核心
2020年第10期24-29,共6页
Microelectronics & Computer
基金
国家973计划项目(2013CB328903-2)。
关键词
小波阈值函数
峰值信噪比
均方根误差
自适应中值滤波
wavelet threshold function
Peak signal-to-noise ratio
Root mean square error
Adaptive median filter