A sweeping fingerprint verification system using the template matching method

Electronic products have adopted the technology of the fingerprint recognition system to offer protection mechanism. The characteristics of Line-Sensor are smaller size, lower power consumption, and lower cost. Therefore, it is suitable for designing fingerprint sensor in embedded systems. Generally, a Line-Sensor builds up a completed picture through assembling the image segments received after one sweeping the fingerprint. However, sometimes, the sweeping rate is unstable and the pressing strength is varied. To avoid such problems influencing the final results, the Fast Normalized Cross Correlation (FastNCC) algorithm is employed here to process and correct the image data. Moreover, the Group Delay Spectrum (GDS) and Dynamic Programming (DP) are implemented to complete fingerprint comparisons. Two different fingerprint matching algorithms are to exploit and demonstrate the verification rate of the system. FastNCC results in 93.8% of the verification rate, while SSD obtains 92.5% of the verification rate.