Title: Content-based image retrievals based on generalised gamma distribution and relevance feedback mechanism
Authors: T.V. Madhusudhanarao; S. Pallam Setty; Y. Srinivas
Addresses: Department of Computer Science and Engineering, T.P. Institute of Science and Technology, Bobbili, 535558, Andhra Pradesh, India ' Department of Computer Science and Systems Engineering, Andhra University, Visakhapatnam – 530003, Andhra Pradesh, India ' Department of Information Technology, GITAM University, Visakhapatnam, 530045, Andhra Pradesh, India
Abstract: The modern technological developments have made significant strides in the usage of internet and made it possible to store and access voluminous data across the globe. These developments have brought certain disadvantages with regard to retrieval of images of interest and also while storing the images. These retrievals play a crucial role in the areas like content-based image retrievals (CBIR). CBIR has wide advantages in areas like security, medical domain where based on the features, the images are to be retrieved. It is therefore necessary to retrieve the interest of images based on certain keyword or features which helps towards faster retrieval of the images. Since the number of images available in the internet is increasing, it is therefore necessary to extract the images based on relevance and query. This paper mainly focuses on the retrieval of the images based on generalised gamma distribution (GGD). The technique employed in this paper is based on the input query and extracting the vital information based on the relevancy. The performance of the developed model is tested using metrics like precision and recall. The methodologies also compared with the available existing methods based on GMM and skew GMM using evaluation metrics. The proposed model exhibited significant performance.
Keywords: content-based image retrieval; CBIR; relevance feedback; query images; generalised gamma distribution; GGD; precision; recall; evaluation metrics.
International Journal of Computational Vision and Robotics, 2015 Vol.5 No.3, pp.271 - 281
Received: 05 Nov 2014
Accepted: 18 Nov 2014
Published online: 20 Aug 2015 *