Applications of Computational Verbs to Effective and Realtime Image Understanding

An effective and realtime image understanding algorithm must meet the challenges of ever-growing image databases, the varieties of features to be understood and the limited human resources for developing this algorithm. In this paper, the reasons that mainstream image understanding algorithms could not meet these challenges were presented. One effective and realtime image understanding platform based on physical linguistic(PL) network and cognitive features was given. The effective aspect of the PL network-based image understanding engine stems from the infinite flexibility of human natural languages when cognitive features are grouped and modelled. The realtime aspect of the PL network-based image understanding engine originates from the much faster processing speed for computational nouns and computational verbs than the processing speed for raw image data. The applications of this new effective and realtime image understanding platform to image search engine, porn detection/filtering and many other image understanding tasks were compared with the conventional approaches. The results show that the new image understanding platform is much more flexible, accurate, user-friendly and less resource consuming. Copyright c © 2006 Yang’s Scientific Research Institute, LLC. All rights reserved.