Image Classification in CBIR Systems with Color Histogram Features

Abstract

Content based image retrieval (CBIR) refers to the ability to retrieve images on the basis of image content. In our work, we describe an approach to CBIR for various database images that relies on human input machine learning and computer vision. More specifically we apply expert level human interaction for solving that aspect of the problem and we employ machine learning algorithms to allow the system to be adapted to new image domains. We present empirical results for the domain of high resolution computed image of flowers. Our results illustrate the efficacy of loop approach to image characterization and the ability of our approach to adapt the retrieval process image domain through the application of machine learning algorithms.