The Research On The Technology Of Divisional Color Histogram Matching And R-tree Organizing Based Image Retrieval

As With the rapid development of the Multimedia technology and Internet,theimage data are sharply increasing. How to get the needed image information from theimage data obtained from satellite system,monitoring system,science test,biomedicalfield and so on is an urgent and important problem.The traditional informationretrieval methods which based on the value/charater can not objectively reflect thediversity of image content,and the data model,system architecture,query method anduser interface based on the traditional methods do not have the capabilities ofmanaging and retrieving image data.So,the studies on the method and technology ofcontent-based image retrieval (CBIR) and image database management have importanttheoretical and practical significance.This paper investigates the content-based image retrieval method from fiverespects including system architecture of CBIR,image data model,orgnigation of imagedata,feature extraction and similar measurement. The result includes two.Firstly,thesystem architecture of CBIR which has wide application nowdays is used.Then a datamodel with image character based on the extended relation data model is proposed.Inaddition,the organigation method of image data is given, and the R-tree is been used asthe index of image database.On the basic of the index method, a K-NN retrievalalgorithm is presented by calculating the MinMax distance. In order to test theperformance of the key technology presented by this paper,a prototype of CBIR systemhas been designed and implemented.The experimental results show that it has betterrecall rate and precision rate than the traditional CBIR systems。...