Shape-based Image Retrieval Using Wavelet Analysis

QBIC (query by image content) which becoming more and more important technique in large image database organization and management is the combine of the image processing and database technique. It is a new research topic was focused in the field of image database technology. Wavelet analysis is internationally recognized up to the minute tool for analyzing time frequency. Based on the work of study and research, the dissertation discusses the method of shape-based image retrieval based on wavelet transform. Image processing is related with image retrieval. It includes image smoother based on two-dimensional discrete orthogonal wavelet transform, image enhancement based on sloped-facet model and the contrast enhancement technique, image enhancement based on fuzzy technique, image segmentation based on edge growing approach. Shape is the important feature of object in image and shape-based image retrieval has more and more attention in recent research on content-based management and utilization of image database system. Although several system has been developed, two main shortcoming are still exited: not stable and the variance with respect to transform, scaling, and rotation. To cure this problem, this paper presents a new method. Though Wavelet Transform get multi-scale edge images, each image is characterized by a multi-scale moment vector in features space. Similarity is given by the Euclidean distance between two images?normalized moment vectors. shape-based image retrieval will be more attention in image database system and will be promising.