Color Image Classification Using Support Vector Machines  

作　　者：冯霞 

机构地区：[1]School of Computer Science and Technology,Civil Aviation University of China,TianJin300300,China

出　　处：《中国民航学院学报》2003年第A02期179-185,共7页Journal of Civil Aviation University of China

摘　　要：An efficient method using various histogram-based(high-dimension al)im age con tent descriptors for automatically classifying general color photos into relevant cate gories is pre-sent ed.Principal component analysis(PCA)is used to project the original high dimensional his tograms onto their eigenspaces.Lower dimensional eigenfeatures are then used to train sup-port vector machines(SVMs )to classify images into their cate gories.Experimen tal results show that even though different descriptors perform differently,they are all highly redundant.It is shown that the dimension ality of all these descriptors,re gard less of their performances,can be signifi cant ly reduced without affecting classification accura cy.Such scheme would be useful when it is used in an interactive setting for relevant feedback in con tent -based image re-trieval,where low dimen sional content descriptors will enable fast on line learn ing and re clas-sification of results.An efficient method using various histogram-based(high-dimension al)im age con tent descriptors for automatically classifying general color photos into relevant cate gories is pre-sent ed.Principal component analysis(PCA)is used to project the original high dimensional his tograms onto their eigenspaces.Lower dimensional eigenfeatures are then used to train sup-port vector machines(SVMs )to classify images into their cate gories.Experimen tal results show that even though different descriptors perform differently,they are all highly redundant.It is shown that the dimension ality of all these descriptors,re gard less of their performances,can be signifi cant ly reduced without affecting classification accura cy.Such scheme would be useful when it is used in an interactive setting for relevant feedback in con tent -based image re-trieval,where low dimen sional content descriptors will enable fast on line learn ing and re clas-sification of results.

关 键 词：支撑向量机 彩色图象 图象分类 PCA SVM 主分量分析 

分 类 号：TN911.73[电子电信—通信与信息系统]