Semi supervised soft label propagation algorithm for CBIR

The increase popularity of using huge image databases in various image retrieval applications construct a need to develop an efficient and robust system provides the output in the form of similar images with respect to the input or query image. Also to carry out its management and retrieval, Content-Based Image Retrieval is an effective method in retrieval system, as well as key technologies. In addition to Compare the shortcoming feature is used in the traditional system, this paper introduces a method that combines color, texture and shape for image retrieval and shows its advantage. But dealing with pattern recognition and machine learning a major problem occur on dimension. Among the dimensionality reduction techniques, Linear Discriminant Analysis is one of the popular methods, but LDA neglect the unlabeled samples. The proposed method propagates the label information from the labeled set to unlabeled set with label propagation process and compare with robust local binary pattern. Further the extensive simulations are conducted on two datasets are Corel 1000 and Brodatz texture database under different noise condition. The results after investigation show a significant improvement in terms of average retrieval precision (ARP) and average retrieval recall (ARR) as compared.