Adaptive Neuro-Fuzzy System based Color Closeness Evaluation

The selection of the proper features and criteria play important role in color processing, segmentation, and recognition systems. The known methods of the color classification mostly rely on empirically defined formulas which do not satisfy all application areas due to the subjective nature of the human vision. This paper suggests a color closeness evaluation system that based on Adaptive Neuro-Fuzzy Inference System(ANFIS). The suggested approach allows to specifically gauge the closeness according to the application area or the particular expert. The feasibility of the method is tested on graphical images selected from BSCS500 database.