Aimed at the limitations of existing keywords-based image search engines on Internet, in this paper, a set of solution based on vision features image search engine is presented. At first, referring to the universal system design norm provided in the MPEG-7, the methods for image features description, extraction and index, high effect algorithms for image features similarity measure and fast retrieval are deep researched, and a new representation combined wavelet with relative moments is given. Then, the advantages of artificial intelligence, data mining and optimal information search strategy on Internet are availably used for constructing a prototype system based on vision features image search engine. The experimental results show that the solution is reasonable and feasible.
Proc. SPIE. 4875, Second International Conference on Image and Graphics
KEYWORDS: Content based image retrieval, Digital signal processing, Digital image processing, Databases, Image processing, Feature extraction, Digital image correlation, Image retrieval, Embedded systems, Communication engineering
Aimed at the limitations existed in content-based image retrieval systems for optimization model, generalization design and retrieval efficiency, in this paper, first, based on the representative method for accumulative histogram on color and co-occurrence matrix on texture, according to the interactive relevance feedback principle between man and computer, a new description method for combined features of image is given. Then, referring to the universal system design norm provided in the MPEG-7, an optimization model of image features correlation retrieval is presented, and a content-based image features correlation retrieval system embedded with DSP is masterly constructed. In this system, the advantages of digital signal processor (DSP) in image signal processing are full utilized, and the parallel processing flows for correlation description of image features and fast retrieval, extraction and index of image features, as well as management and renewal of image data are availably performed, which brings forward a new idea for appropriative image retrieval in large image database (such as remote sensing images in GIS) and universal image retrieval in Web database (such as different images on Internet). The rationality of the new scheme is validated by experimental results.