摘要
为了能在去除图像噪声的同时有效地克服Gibbs现象,得到令人满意的视觉效果,提出了一种基于局部自适应阈值的小波图像降噪方法.该算法利用局部化信息和层间相关性理论,对小波系数进行分块分类处理.该算法首先把图像划分成子块,通过调节全局阈值得到各个子块阈值,从而有效地利用了局部信息,有选择地对图像进行降噪处理.算法加入自适应的步骤,对于不同尺度的子带,分别赋予大小不同的阈值,使算法具有更好的自适应性.试验结果表明,与其他几种传统降噪方法相比,该方法能获得较好的降噪效果.
To effectively eliminate image noise and overcome the Gibbs defect,a wavelet image denoising method based on local adaptive threshold is proposed in this paper.Based on the local information and interscale dependency of wavelet coefficients,the wavelet coefficients are split respectively.The algorithm makes use of the local information selectively to denoise by splitting image into small blocks and adjusting the global threshold to get the local ones.Considering an adaptive step into the method to improve the algorithm＇s adaptability,we make a difference on the subbands＇ thresholds related to different scales.Experimental results show that this method can obtain better denoising performance in comparison with common methods.
出处
《青岛理工大学学报》
CAS
2010年第2期118-122,共5页
Journal of Qingdao University of Technology
基金
青岛大学青年科研基金项目(2007005)
关键词
小波分析
图像降噪
局部信息
自适应阈值
wavelet analysis
image denoising
local information
adaptive threshold
