DSP based embedded fingerprint recognition system

• Published in: 13th International Conference on Hybrid Intelligent Systems (HIS 2013) pp. 6 - 11
• Main Authors: Kamaraju, Maddu, Kumar, Penta Anil
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2013
• Subjects: Biometric; Clocks; Digital signal processing; DSP processor; Feature extraction; Fingerprint; Gabor filter; Image enhancement; Image recognition; Image segmentation; Ports (Computers); Recognition
• DOI: 10.1109/HIS.2013.6920460

Along with the rapid development of biometric recognition techniques, the fingerprint recognition became a significant subject. Fingerprint recognition is a method to identify a person, based on the physiological characteristics of the finger. Many of today's recognition systems are realized on the PC-based platform, which has high power consumption and cannot be used with ease. But, the recently developed market for civil applications has more demand for system's portability and power consumption. So, the trend of miniaturizing and embedding the system for automatic fingerprint identification became more distinct. In this paper, a convenient, fast, and inexpensive low power Digital Signal Processor (DSP) based embedded fingerprint recognition system is introduced. Reliable extraction of features from input fingerprint image is the most challenging problem faced in the area of fingerprint recognition. However the performance of the Minutiae extraction algorithm relies heavily on the quality of the input fingerprint image. In order to ensure to extract the true minutiae points it is essential to incorporate a good enhancement algorithm. The earlier fingerprint image enhancement methods are FFT based and has a drawback of poor image quality which results in unreliable minutiae extraction and thereby reduces the accuracy of recognition result. In this paper, at the algorithm level a new approach for fingerprint image enhancement based on the Gabor filter is introduced. Comparatively this algorithm produced the good results in the view of image quality and accuracy.