摘要
利用小波变换在时、频域同时具有良好的局部化性质,提出由小波变换得到的多分辨率表示来进行纹理识别.通过对肝脏B超图象进行分类,结果表明,本文提出的方法具有分类精度高的特点.
Acoording to the fact that wavelet transform remains well localized in hath time spatial and frequency domains this paper presents a method of texture classification by multiedution representations obtained ho wavelet transform . The method is applied to the classification of ultrasonic liver images and produces about 95% of comet classification rate.
出处
《海南大学学报（自然科学版）》
CAS
1997年第4期272-275,共4页
Natural Science Journal of Hainan University
