Color recognition of license plates using fuzzy logic and learning approach

To address the changing illumination conditions and different color features in different places,an algorithm for color recognition of license plates using fuzzy logic and learning approach is proposed.The HSV(hue,saturation and value)color space was employed to perform color feature extraction.The 4-class problem is converted to be two 2-class problems based on reverse color information.The classification function was built on the integration of fuzzy maps on the respective sets of three components in HSV space.To improve the adaptability,the correlative parameters were obtained by a learning algorithm.Experiments were conducted on a DSP-based embedded license plate recognition platform.The results show that the proposed algorithm achieves better recognition performances on two test sets.compared with some traditional classification approaches.