摘要
针对传统Bayes阈值不能随小波分解尺度变换以及提高传统算法图像降噪效果的问题,文章提出一种改进的基于小波维纳滤波与Bayes自适应阈值估计图像降噪算法,该算法在多层小波变换的基础上,对小波分解后的第一层细节系数进行维纳滤波处理,对其他层细节系数进行改进Bayes软阈值估计算法处理,最后对处理后的小波系数进行重构,得到降噪图像。实验结果表明,该方法在图像峰值信噪比(PSNR)定量指标上优于传统的小波Bayes软阈值估计图像降噪方法,并将该方法成功的应用于轴承缺陷图像的降噪预处理以及轴承缺陷图像边缘检测中,达到了图像降噪的优化效果。
Aiming at the problems that the traditional Bayes threshold can not be changed with the wavelet decomposition scale and the image noise reduction effect of the traditional algorithm is improved. An improved image de-noising algorithm based on wavelet wiener filtering and Bayes adaptive threshold estimation is proposed in this paper. Based on the multilayer wavelet transform, the first layer detail coefficient is pro- cessed by wiener filtering, and the other layers detail coefficients is processed by the improved Bayes soft threshold estimation method. Finally, the processed wavelet coefficients are reconstructed to obtain the denoising image. The experimental results show that this method is superior to the traditional wavelet Bayes soft threshold estimation image denoising method in the image peak signal to noise ratio （PSNR） . The method is successfully applied to beating defect de-noising image and edge detection. Finally, the optimization effect of de-noising image is achieved.
出处
《组合机床与自动化加工技术》
北大核心
2017年第11期65-68,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家高技术研究发展计划(863计划)(2015AA043702)
北京市教委科技计划重点项目(KZ201611232032)
北京市教委科研计划项目(KM201611232020)
关键词
小波分解
维纳滤波
阈值估计
图像降噪
峰值信噪比
wavelet decomposition
wiener filtering
threshold estimation
image de-noising
peak signal to noise ratio
