A FAST AUTOMATIC FINGERPRINT IDENTIFICATION METHOD BASED ON A WEIGHTED-MEAN OF BINARY IMAGE

This paper first proposes a fast fingerprint identification method bas ed on a weighted-mean of binary image, and further investigates optimi zation of the weights. The proposed method uses less computer memory t han the conventional pattern matching method, and takes less computati on time than both the feature extraction method and the pattern matchi ng method. It is particularly effective on the fingerprints with a sma ll angle of inclination. In order to improve the identification precis ion of the proposed basic method, three schemes of modifying the propo sed basic method are also proposed. The performance of the proposed ba sic method and its modified schemes is evaluated by theoretical analys is and computer experiment using the fingerprint images recorded from a fingerprint read-in device. The numerical results showed that the pr oposed method using the modified schemes can improve both the true acc eptance rate and the false rejection rate with less memory and complex ity in comparison with the conventional pattern matching method and th e feature extraction method.