Authors : Vishal B. Padole and V.R. Mankar
Abstract: We formulate the problem of using an individual classifier for searching the images in the huge database by providing image or its rough sketch as a query. In order to used single classifier for image classification it produce numerous drawbacks, so to avoid this problem a systems comprising of a pool of classifier has to be designed which generates the output which is complement of each other. Though this field has acquired much wide popularity, CBIR has not yet reached its mellowness and is not yet being used on a large scale. In this study, various concepts regarding CBIR like objectives while designing a CBIR system, problems associated while designing an efficient CBIR system, analysis of KNN classifier for this system and a new system overcoming the drawback of using existing methods in terms of computational time and correctness have been discussed.
Vishal B. Padole and V.R. Mankar, 2017. Content Based Image Retrieval System Based on Classifier Ensemble Approach. Journal of Engineering and Applied Sciences, 12: 9415-9418.