Research And Implementation Of The Multiple Method To Improve The Recognition Rate Of Fingerprint

With the rapid progress of the technology and computer techniques, informatization and intellectualization have been the mainstream of the society. Therefore, the information security issues becomes more important. The fingerprint identification technology is widely used thanks to its high security, convenience and low cost. Now it has shown long-term development in the area of attendance access control, safe deposit box, door lock and cellphone. Although this technology has been widely used in the inland, the drawbacks of itself and the influence of external factors lead to the low rate of fingerprint recognition.Aiming at the problem of low identification rate, with the practice in the company, the technology of fingerprint that i've mastered and the current technology of fingerprint identification, this article provides a new multi-templates method to improve the rate of identification. To update the fingerprint template, we make the templates of the fingerprint added in the phases of search and comparison. And this method is put forward towards to the problem of low rate of fingerprint identification of children, the old, and even caused by the fingerprints are not clear and so on.Our research propose a multi-model Fingerprint recognition method and the related implementation. To overcome the low rate of single-model Fingerprint recognition, we propose the multi-model Fingerprint recognition method instead. In this work, we build the flowchart of Fingerprint registration and fingerprint identification by analyzing multi-model Fingerprint recognition, and experimentally verify our method in Meta Ware. we run a test on hardware platform as well, draw our conclusion by analyzing all the test data. Our research include a detailed analysis of Pre-processing fingerprint algorithm and optimization of Image Enhancement Algorithm as well.There little difference of the Auto Sensing between the Multi-template method and single template fingerprint identification system when dealing with the normal finger with better quality. However, when it comes to those of the kids and aged with bad lines, the Auto Sensing of Multi-template method is higher than the one of single template fingerprint identification system. And for the hybrid fingerprint, it also outperforms the single template fingerprint identification system. As a whole, the recognition rate of this proposed method is 40% higher than the one of single template fingerprint identification system. Furthermore, in this method, the fingerprint templates are updated during the stage of searching and contrasting. The fingerprint register stage will not occupy much time of the users. The fingerprint stitching algorithm is applied in fingerprint algorithm. The fingerprint products which use this method adapt to the market, making the fingerprint identification technology improve the living quality of more people and booming the economic development of the Fingerprint identification industry.