Semantic Image Analysis for Intelligent Image Retrieval

Abstract Image understanding and analysis is the most exciting and fastest-growing research areas in the computer vision. Recent computer vision technologies and algorithms are support efficient semantic image analysis and retrieval. Image analysis is deal with image representation, estimation formula, and sampling density. Image analysis at semantic level is result in automatic extraction of image descriptions as per human perception which ultimately bridge semantic gap between low-level visual features and the high-level concepts capturing the conveyed meaning. Vital semantic image information is basically retrieved from image content, mainly from meaningful image objects and their mutual relations. In this paper, we present Semantic analysis of image by knowledge driven approach, start with Image content analysis with respect to semantic concepts, design image database and knowledge base on the basis of semantic content and retrieval, presentation and modification of image reference database or knowledge base for knowledge delivery intention. Experimental result shows improvement in image retrieval performance and accuracy.