Fingerprint Verification Locator Embedded Software Development And Key Technology Research

With fingerprint acquisition sensors becoming more lightweight and miniaturized,fingerprint recognition technology has been widely used in the field of authentication for mobile devices.Fingerprint recognition solutions based on fingerprint recognition modules occupy the mainstream of the current market due to their short development cycle.In this thesis,a fingerprint verification locator is developed based on fingerprint recognition module with the background of community correction supervision work,and the fingerprint recognition system and fingerprint recognition algorithm of the scheme are optimized.The main research work and innovation points are as follows:(1)Based on the application background and actual needs of community correction projects,the basic functions of the fingerprint verification locator are clarified,a fingerprint verification locator fingerprint authentication scheme based on fingerprint recognition module is proposed,and the hardware and software development platform selection is completed through the analysis of the functional requirements of the device.(2)In order to shorten the development cycle,this thesis firstly takes MT2503 as the development platform and combines it with fingerprint recognition module to realize the fingerprint verification locator.At the same time,an improved scheme of fingerprint recognition system based on fingerprint acquisition sensor is proposed to transfer the acquired fingerprint images to MT2503 through SPI interface for relevant algorithm processing,which further improves the real-time performance of fingerprint matching and hardware utilization of the device.The experimental results show that the average time required for fingerprint matching in the improved scheme is reduced by 34 ms compared with the original scheme,and the hardware cost is greatly reduced.(3)To address the problem that the traditional fingerprint recognition algorithm based on detailed feature points has a high rejection rate when dealing with small-area fingerprint recognition,this thesis proposes a small-area fingerprint recognition algorithm based on the improved ORB algorithm.The algorithm reduces the mis-matching rate of ORB algorithm by introducing LBP descriptors and cosine similarity,and then uses the improved ORB algorithm to realize small-area fingerprint splicing to expand the effective registration area and reduce the rejection rate of smallarea fingerprint recognition.The experimental results show that the rejection rate of the method is 6.4%,the false recognition rate is 0.1%,and the average matching time of a single fingerprint is 58 ms on the self-built small-area fingerprint database,all of which are better than the traditional fingerprint recognition algorithm and can effectively improve the fingerprint recognition performance of the device.In this thesis,the fingerprint verification locator is tested for practical application in combination with the community correction positioning platform.The test results show that the device’s functions and performance indicators are in line with the expected objectives and meet the actual application requirements.At present,the device has been practically applied to the supervision of community corrections in Hunan Province,and more than 1000 correctional officers have been using this device.