ABSTRACT

Content-based image retrieval (CBIR) systems normally return the retrieval results according to the similarity between features extracted from the query and candidate images. In certain circumstances, however users are most qualified to specify the query “content” or objects(e.g., Eiffel Tower) of their interest , not the computer and only wish to retrieve images containing relevant objects, while ignoring irrelevant image areas (such as the background). Previous work on this normally requires complicated segmentation of the object from the background. In this paper, the user can select “object of user’s interest” of different shapes, non homogenous texture containing different colors regardless of many objects present in the same image using varied tools like polygonal, rectangle, circle selector tools. A twostate procedure is used to query the image from the Image database. First, we integrate global color and texture feature vectors to narrow down the search space and in next state we Process using local features. We use color moments and subband statistics of wavelet decomposition as color and texture features respectively. The shape features, generated by mathematical morphology operators, are further employed to produce the final retrieval results.