基于小波变换的图像阈值化去噪算法研究  

Research on Image Thresholding Denoising Algorithm Based on Wavelet Transform

作　　者：刘建南 

机构地区：[1]吉林省测量标志管理站

出　　处：《测绘与空间地理信息》2009年第4期105-107,111,共4页Geomatics & Spatial Information Technology

摘　　要：鉴于传统的去噪方法难以在各频域得到很好的兼顾,本文提出一种基于小波变换的去噪方法。通过对图像进行2维离散小波变换(DWT),再根据噪声和图像信号的不同特性,在小波域中,选取适当的阈值对小波系数进行处理,最后进行小波重构(IDWT),得到去噪后的图像。实验表明,该算法具有很好的效果。According to traditional denoising methods which are difficult to take account of the timefrequency domains, this paper put forward a denoising method based on wavelet transform. Based on 2D Discrete Wavelet Transform（DWT） for image, then aiming at the different character of noise and image signal, in the domain of wavelet, it processed the wavelet parameters using reasonable threshold value. Finally, it acquired the image after denoising by inverse discrete wavelet transform （IDWT）. The experiments show that the algorithm has better effects.

关 键 词：小波变换 图像去噪 阈值化 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]