A Study Of Fingerprint Image Acquisition And Processing Based On DSP

With the development of biometric recognition technology, such as recognition of fingerprint, palm, voice, iris, face, etc., fingerprint identification has become a focus of biologic identification and been used in many fields, because of its accuracy, reliability and convenience. This article presents the research of a fingerprint identification stand-alone system, aiming to widen its applications and promote its further development. The system is composed of micro fingerprint grabber, image acquisition card, microprocessor, DSP, CPLD and identification module. It can be used in fingerprint lock, lockfast, time attendance and so on.To meet the needs of modern techniques and the system realization, the structures of the system are designed properly and the devices in the system are selected carefully. More over, both the cost and the quality are also considered. The main hardware of the system consists of MCS51 microprocessor, MAX7000S of Altera, SAA7110 of PHILIPS, and fingerprint grabber. The main integrated software consists of algorithm based on TMS320C54x of TI.This paper includes four sections. Section one presents real-time fingerprint image acquisition, including dynamic adjust of brightness and contrast for clearness improvement. Section two describes imageenhancement to improve the image quality for feature extraction. Section three addresses fast feature extraction and classification of fingerprint image based on advanced DSP resolution. The last section states the experiments, global system tests and results.The results show that the stand-alone system connecting with fingerprint lock works well. It also shows that the design of the system is reasonable and realizable. The system has the advantages of easy operation, easy update, high reliability .high speed and low cost. It can be used in fingerprint identification system or be embedded in other systems, It would have a wide range of applications and a bright future.