摘要
利用对原始乳腺图像进行小波变换后得到的子图来检测肿块．肿块在数字乳腺图像频域上的表现明显异于周围区域．使用适当的小波变换能够把这种特征在变换域中比较直观地显示出来．为了检测出肿块位置，利用Biorthogonal (bior Nr.Nd)小波函数族对３９幅原始乳腺图像进行小波变换，分解为包含不同频带的子图．分析子图的结果表明：这种基于小波变换的方法可用于检测数字乳腺图像上的肿块．
This paper presents an approach for analyzing masses in digital mammograms employing wavelet-based subband image decomposition. In the frequency domain the masses show relative difference from their neighboring region. These image features can be preserved by an analyzing system that employs a suitable image transform, which can localize the signal characteristics in the transform domain. Detection of masses is achieved by decomposing 39 mammograms with biorthogonal (biorNr. Nd) wavelets into different frequency subbands. The results show the potential of wavelet-based subband image decomposition as a tool for detecting masses in digital mammograms.
出处
《上海大学学报（自然科学版）》
CAS
CSCD
2000年第6期538-540,547,共4页
Journal of Shanghai University:Natural Science Edition
基金
上海市教委发展基金!(97A45)
