Performance Evaluation of Ontology and Fuzzybase CBIR

Abstract

IN THIS PAPER,WE HAVE DONE PERFORMANCE EVALUATION OF ONTOLOGY USING LOW-LEVEL FEATURES LIKECOLOR,TEXTURE AND SHAPE BASEDCBIR,WITH TOPIC SPECIFICCBIR.THE RESULTING ONTOLOGY CAN BE USEDTO EXTRACT THE APPROPRIATE IMAGES FROM THE IMAGE DATABASE.RETRIEVING APPROPRIATE IMAGES FROM ANIMAGE DATABASE IS ONE OF THE DIFFICULT TASKS IN MULTIMEDIA TECHNOLOGY.OUR RESULTS SHOW THAT THEVALUES OF RECALL AND PRECISION CAN BE ENHANCED AND THIS ALSO SHOWS THAT SEMANTIC GAP CAN ALSO BEREDUCED.THE PROPOSED ALGORITHM ALSO EXTRACTS THE TEXTURE VALUES FROM THE IMAGES AUTOMATICALLYWITH ALSO ITS CATEGORY(LIKE SMOOTH,COURSE ETC)AS WELL AS ITS TECHNICAL INTERPRETATIO