The Research And Implementation Of MPEG-7 Retrieved System

Recent years have witnessed the rapid growth of digital images due to the increasing power of computing and the fast development of Internet. How to retrieve the images in which people are interested from those is becoming an issue which needs solving urgently. Content-based Image Retrieval (CBIR) is the key technology to address this urgent issue. CBIR is a technology for retrieving images on the basis of image features such as color, texture and shape, etc. However, the content features of images are complicated and different at all, which make the retrieval efficiency of CBIR systems lower. According to this, MPEG (Moving Picture Experts Group) of ISO/IEC developed the ISO/IEC standard MPEG-7.In this paper, the main objective aims at the image retrieval based on MPEG-7. How to pick up the shape descriptor from MPEG-7 is studied thoroughly and some retrieval experimentations and performance analysis on regional shape descriptor is done. DART method based on regional shape ART arithmetic in MPEG-7 is put forward, which is used in discrete digital image implementation. One picking-up method ARS shape descriptor is put forward. The retrieving results of ARS shape descriptor are far better than DART which is based on regional shape too. Finally a practical image retrieval system is developed by using regional shape.