基于小波包分析的颅颌面纹理特征提取方法  

On Cranio-Maxillo-Facial Texture Feature Extraction Algorithm Based on Wavelet Packets Transform

作　　者：黄丽雯[1] 庞柯[1] 汪鑫[1] 施帮利[1] 王涛[2] 炊万年[1] 

机构地区：[1]重庆理工大学电子信息与自动化学院,重庆400054 [2]重庆医科大学附属口腔医院,重庆口腔病及生物医学科学研究中心,重庆401147

出　　处：《西南师范大学学报（自然科学版）》2013年第11期59-63,共5页Journal of Southwest China Normal University(Natural Science Edition)

基　　金：重庆市自然科学基金计划资助项目(CSTC;2010BB2311)

摘　　要：针对新型生物研究对象颅颌面的研究,提出小波包分析的方法对颅颌面的纹理特征进行提取.特征矢量为颅颌面预处理后图经过小波包分解后各个子节点的节点能量占总能量的比例.利用该特征矢量,计算相似度,实验证明,该方法计算简单,特征维数较低,可以有效提取颅颌面的纹理特征.A usage of wavelet packets transform has been proposed to extract the cranio-maxillo-facial texture feature to research this new kind of Biological object. Characteristic vector is the wavelet tree node energy, got by preprocessing and wavelet packets decomposition. With this characteristic vector, the similarity of different cranio-maxillo-facial image has been calculated. Experiment shows that the method features in simple calculation and low dimensions, which is capable of extract the cranio-maxillo-facial texture feature effectively.

关 键 词：颅颌面 纹理特征 小波包 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]