Abstract
The retrieval of images using semantic concepts is challenging and essential issue in watchword based search. The focus has been shifted from low-level feature extraction to human semantics in order to enhance the precision of content based image recovery framework. It is done to decrease the semantic gap between visual components and affluence of human semantics. This research provides an ontology-driven approach to train the system by modeling the human cognition that also conquers the constraints of a label based retrieval. In the proposed work, a properly characterised philosophy framework is used to drive image annotations that enable semantic retrieval of aerial photography. The concept of image semantics is exploited for the classi¬fication of vegetation and also to determine the qualitative semantics by using Allen’s qualitative relations. Psychophysical evaluation is used to assess the effectiveness of the proposed approach. The outcomes of the different experiments are relatively favorable in terms of accuracy of relevant image retrieval.