A Fuzzy Neural Network Approach for Document Region Classification Using Human Visual Perception Features
Chacón Murguía, Mario Ignacio · 2002
THIS PAPER DESCRIBES A FUZZY NEURAL NETWORK CLASSIFIER TO PERFORM DOCUMENT REGION CLASSIFICATION USING FEATURES OBTAIND FROM HUMAN VISUAL PERCEPTION THEORIES. THE FOUNDATIONS OF THE CLASSIFIER ARE DERIVED FROM HUMAN VISUAL PERCEPTION THEORIES. THE THEORIES ANALYZED ARE TEXTURE DISCRIMANATION BASED ON TEXTOS, AND PERCEPTUAL GROUPING. BASED ON THESE THEORIES, THE CLASSIFICATION TASK IS STATE AS A TEXTURE DISCRIMINATION PROBLEM AND IS IMPLEMENTED AS A PREATTENTIVE PROCESS. ENGINEERING TECHNIQUES ARE THE DEVELOPED TO EXTRAC FEATURES FOR DECIDING THE CLASS OF INFORMATION CONTAINED IN THE REGIONS. THE FEATURE DERIVES FROM THE HUMAN VISUAL PERCEPTION THEORIES A MEASUREMENT OF PERIODICITY OF THE BLOBS OF TEX REGIONS. THIS FEATURE IS USED TO DESIGN A F