Semantic Image Retrieval in a Grid Computing Environment Using Support Vector Machines

摘要

In this paper, we propose a multiple support vector machine-based architecture for content-based image retrieval (CBIR) in a grid computing environment In order to maximize the performance of the proposed technique, an efficient feature extraction method is introduced, which is based on the concept of in-depth texture analysis. For this, we are using wavelet packets, Gabor filters and curvelet transformed features for the repository image representation. To ensure semantically identical image retrieval, an association scheme is presented which utilizes OurGrid computational grid, and guarantees the retrieval of images in an efficient way. To demonstrate the effectiveness of the present work, the proposed method is compared with several existing CBIR systems, which shows that the proposed method performs better than all of the comparative systems.