摘要
针对曲波降噪方法在获得高质量图像的同时会出现一些虚像的问题,提出了一种用来抑制虚像的基于模糊划分的降噪算法。算法基于结果图像的曲波和隐藏马尔科夫树的降噪方式设计,图像经过模糊划分后,模糊窗口的性能可以估算,而基于曲波和小波结果图像的图像融合权值可以由模糊窗口的性能确定。试验结果证明,文中算法能有效地提高结果图像的视觉效果并能明显地抑制结果图像的虚像。
The curvelet denoising can obtain high quality image,however some virtual images may occur in the result,hereby a image denosing algorithm based on fuzzy partition was proposed to overcome this problem. The algorithm is designed based on curvelet and Hidden Markov tree denoising for the result images,the fuzzy window performance can be estimated after fuzzy partition,and the fusion weights of curvelet and wavelet can be determined by the performance of the fuzzy window. The image processing results show the effectiveness of the proposed algorithm which can effectively improve the image quality and suppress the virtual image.
出处
《数字通信》
2014年第6期25-27,共3页
Digital Communications and Networks
关键词
曲波降噪
模糊划分
隐藏马尔科夫树
小波变换
curvelet denoising
fuzzy partition
Hidden Markov tree
wavelet transform
