Abstract: CBIR relies on the searches purely is based on metadata obtained from the images like features and annotation information. The evaluation and the effectiveness of image search is more important and has been well-defined. Users of image databases often prefer to retrieve relevant images by categories. Unfortunately, images are usually indexed by low-level features like color, texture and shape, which often fail to capture high-level concepts well. To address this issue, relevance feedback has been extensively used to associate low-level image features with high-level concepts. Among all existing relevance feedback approaches, query movement and feature re-weighting have been proven to be suitable for large-scaled image databases with high dimensional image features. In this paper, we investigated different weight update schemes and compared the retrieval results. As far as feature re-weighting approaches are concerned, one of their common drawbacks is that the feature re-weighting process is prone to be trapped by suboptimal states. To overcome this problem, we introduce a disturbing factor, which is based on the Fisher criterion, to push the feature weights out of sub-optimum. Experimental results show that this method performances well compared to basic re-weighing methods
Keywords: Content based image retrieval, re-weighing features, Short term Learning, Long term learning.