基于多尺度几何分析的目标描述和识别  

Object description and recognition using multiscale geometric analysis

作　　者：潘泓[1] 李晓兵[1] 金立左[1] 夏良正[1] 

机构地区：[1]东南大学自动化学院,江苏南京210096

出　　处：《红外与毫米波学报》2011年第1期85-90,共6页Journal of Infrared and Millimeter Waves

基　　金：国家自然科学基金(90820009;60805002);航空科学基金(20080169003);东南大学优秀青年教师教学科研资助计划(4008001015);国家留学基金委资助计划

摘　　要：结合多尺度几何分析和局部二值模式算子,构造了一种新的多尺度、多方向局部特征描述子——局部Contourlet二值模式(LCBP).通过对尺度内、尺度间及同一尺度不同方向子带内LCBP直方图统计分析,同时考虑到LCBP的四叉树结构特点和模型的简单性,用两状态HMT描述LCBP系数,得到LCBP-HMT模型.在此基础上,提出了基于LCBP-HMT模型的目标识别算法,该算法提取LCBP-HMT模型参数作为特征,通过比较输入目标特征和各类标准目标特征的Kullback-Leibler距离进行分类.实验结果表明,LCBP特征比传统小波域特征和Contourlet域高斯分布模型特征更具鉴别能力.A novel local feature descriptor,called Local Contourlet Binary Pattern（LCBP）,was developed in this paper.LCBP provides a multiscale and multidirectional representation for images since it integrates multiscale geometric analysis and local binary pattern operators.With the quadtree structure of LCBP and simplicity of the model itself,the LCBP coefficients were modeled by a two-state HMT that is in accordance with the intra-band,inter-band and inter-directional distributions of LCBP coefficients.Based on the LCBP-HMT model,an object classification method was further proposed to extract parameters of the LCBP-HMT model as features and classify the query samples by comparing the Kullback-Liebler distance between features of the query samples and that of the prototype objects.Experimental results illustrate the superiority of the LCBP over traditional wavelet features and Gaussian density function model features of contourlet coefficients in terms of the discrimination performance.

关 键 词：多尺度几何分析 轮廓波变换 局部二值模式 目标识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]