Study Of JPEG2000 Image Indexing In Compressed-domain

In recent years, with the development and spread of many compressed standards,compressed domain image retrieval techniques have become the study topics in thecontent-based image retrieval(CBIR) domain. JPEG2000, as the new generation of stillimage compressed standard, supports some very distinguished features and has gettonsuccess in applications in some important fields.In this paper, some of exploratory research work has been done around theJPEG200 image retrieval techniques in the compressed domain. The main work can besummarized as follows:1. A full-scale discussion of the current compressed domain retrieval techniques is made. The features and classification of compressed domain retrieval techniques are introduced and the retrieval methods in the transform domain and spatial domain are fully discussed.2. The general information of the JPEG2000 still image compression standard is introduced, and each part of the JPEG2000 coder is discussed in detail according to the JPEG2000 encoding order.3. A JPEG2000 image indexing algorithm in compressed domain is proposed. On the basis of DWT and EBCOT using in JPEG2000, texture and color features are extracted with high memory efficiency. K-L transform are used to reduce dimensionality. These two features are combined for efficient image retrieval and achieved good results.