Content Based Image Retrieval System Using Relevance Feedback

Content based image retrieval (CBIR) is the basis of image retrieval systems. Image retrieval based on image content has become an interesting topic in the field of image processing. To be more profitable, relevance feedback techniques were introduced into CBIR such that more precise results can be obtained by taking user’s feedbacks. However, existing relevance feedback based CBIR methods usually request a number of iterative feedbacks to produce refined search results, especially in a large-scale image database. To achieve the high efficiency and effectiveness of CBIR we are using two type of methods for feature extraction like SVM (support vector machine) and NPRF (navigation-pattern based relevance feedback). By using SVM classifier as a category predictor of query and database images, they are exploited at first to filter out irrelevant images by its different low-level, concept and key point-based features. In terms of effectiveness, the search algorithm makes use of the discovered navigation patterns and three kinds of query refinement strategies, Query Point Movement (QPM), Query Reweighting (QR), and Query Expansion (QEX) to convert the search space toward the user’s intention effectively. By using these methods, high quality of image retrieval on RF can be achieved in a small number of feedbacks.