摘要
利用小波变换对图像去噪是一种非常有效的方法。传统的小波去噪算法对图像去噪后的平滑效果不是很好,图像细节清晰度不够高,甚至会产生伪吉布斯现象。针对这些现象,文中提出了一种改进的基于小波变换的多尺度自适应阈值图像去噪方法。该方法根据图像小波分解的特性,确定适合小波分解后不同层系数去噪的较优阈值,然后结合恰当的阈值函数对各层高频系数进行处理来达到去噪效果。实验结果表明,与传统方法相比,该方法运算量较小,能有效去除高斯白噪声,进一步提高峰值性噪比,同时能够很好地保留图像细节信息。
Using wavelet transform to filter noises on image is a very effective method. The smoothing effect is not very good of tradition- al wavelet image denoising algorithm, and the image detail precision isn＇ t high enough, even false Gibbs phenomenon can be produced. Aimming at the phenomenon ,an improved multi-scale adaptive threshold method of image denoising based on wavelet transformation has been proposed. According to the characteristics of the image wavelet decomposition, this method can determine the better threshold of dif- ferent layers＇ coefficient for denoising after wavelet decomposition, then process the high frequency coefficient of each layer with appro- priate threshold function to achieve denoising effect. The experimental results show that, compared with traditional methods, this method can effectively remove Ganssian white noise and further improve the peak signal-to-noise ratio,while well preserving image details.
出处
《计算机技术与发展》
2013年第8期250-253,共4页
Computer Technology and Development
基金
江苏省科技计划资助项目(BK2010546)
关键词
图像去噪
小波变换
多尺度
自适应阈值
峰值信噪比
image denoising
wavelet transform
multi-scale
adaptive threshold value
peak signal to noise ratio
