Technology Research, Content-based Image Retrieval

With the development of the multimedia and network technology, the application of the image is extensive and content-based image retrieval technique has already become the studied focus. It has combined the technologies, such as image processing, pattern-recognition, vision understanding, databases,etc. It is an extremely promising new technology in research and application.This paper has introduced the state-of-the-art of the research and application,designed the frame of an image retrieval system, and put forward the data structure model.Shape-based image retrieval is a key technology in content-based image retrieval. There are three important issues involved in shape similarity retrieval: feature extraction,similarity measure and feature matching.On these issues,overall analysis and commentary has been carried on.The course of extracting shape feature is: after medium filtering,edge detection and thresholding,the edge of image is obtained, then it is simplied as approximate polygon.A mutual template matching approach is proposed. (l)Experiment results show that algorithm is invariant with respect to translation, scaling and rotation. (2)Because the optimization method is adopted , this algorithm can realize shape matching of convex polygons with small difference. (3)In addition, this algorithm can match polygons with different sides, so it offers a criterion of measuring similarity of polygons with heavy difference .By observing the procedure of human' s understanding image ,it can be found that: the procedure of shape matching is hierarchical. Exactly based on this idea, sub-model decomposition approach to concave polygon and hierarchical model weighted measure are presented. Using these methods, similarity measure to simple concave polygons has been realized .