基与结合遗传算法结合的直觉模糊集的检索方法(英文原稿).pdfVIP

J ou rnal of B eij ing I ns tit ute of Technology , 2009 , Vol . 18 , No . 1 Image Retrieval Approach Based on Intuitive Fuzzy Set Combined with Genetic Algorithm WAN G Xiaoyin (王潇茵) 1 , 　XU Weihua (徐卫华) 2 , 　HU Changzhen ( 胡昌振) 1 ( 1. Lab of Computer Network Defense Technology , Beijing Institute of Technology , Beijing 100081 , China ; 2 . Automation Station of Army Aviation Dep art ment of t he General Staff , Beijing 1000 12 , China) Abstract : Aiming at shortcomings of traditional image retrieval systems , a new image retrieval approach based on color features of image combining intuitive fuzzy t heory wit h genetic algorit hm is proposed . Each image is segmented into a constant number of subimages in vertical direction . Color features are extracted from every subimage to get chromosome coding. It is considered t hat fuzzy membership and intuitive fuzzy hesitancy de gree of every pixel ’s color in image are associated to all t he color histogram bins. Certain feature , fuzzy feature and intuitive fuzzy feature of colors in an image , are used toget her to describe t he content of image . Efficient combinations of subimage are selected according to op eration of selecting , crossing and variation . Retrieval re sult s are obtained from image matching based on t hese color feature combinations of subimages. Test s show t hat t his approach can improve t he accuracy of image retrieval in t he case of not decreasing t he sp eed of image retrieval . It s mean precision is above 80 %. Key words : intuitive fuzzy ; genetic algorit hm ; color histogram ; image retrieval CLC number : T