A METHOD OF EXTRACTING FEATURES APPLIED FOR CONTENT-BASED IMAGE RETRIEVAL

Abstract

In this paper, we approach a method of extracting for the content based images retrieval. In order to solve this problem, the technics of extracting features are proposed including segmentation of image rely on contrast, extracting the boundary of interest object based on LoG method, extracting texture based on the filter operations, and extracting color histogram rely on the method of clustering the intenses of image’s pixels. On the base of these features, we extracted the feature vector to describle the image content by location, color, shape. From that, a method of similar images retrieval with a query image is built. To prove the proposed theory, an experimental was built on the datasets including COREL, Wang and compared with other works on the same data set to evaluate the accuracy of our proposed method. The experimental results show that our method of extracting image features applied to our similar image retrieval problem is effective