Research And Implementation Of A Fingerprint Identification System Based On FPGA

As the most traditional and mature biometric application scheme,fingerprint feature is widely used in authentication and recognition.With the development of IC industry,portable electronic devices require low power consumption,small volume and high speed for fingerprint authentication system.Nowadays,traditional fingerprint authentication systems based on software and embedded microprocessor are increasingly difficult to solve the execution speed requirements of complex authentication algorithms.Therefore,this thesis implements the fingerprint authentication system in FPGA based on pure hardware algorithm circuit,and accelerates the authentication algorithm by utilizing the powerful parallel computing capability and customizability of FPGA.This scheme is used to solve the performance bottleneck of fingerprint authentication system.This thesis takes the fingerprint authentication system based on FPGA as the research topic,and the main research includes:1.Based on the current basic structure and research status of fingerprint authentication system,taking the algorithm processing flow in fingerprint authentication system as the context,this thesis studies the principle and function of fingerprint preprocessing and identification matching algorithm.The algorithm model is established in MATLAB to preview the execution effect of some algorithms and the whole system.2.According to the characteristics of digital circuit,this thesis selects and improves the algorithm scheme.It focuses on the Gabor filter enhancement algorithm of integer quantization,the Otsu binarization algorithm based on gray histogram,the OPTA thinning and feature point extraction algorithm based on look-up table method,and the feature point matching algorithm based on vector triangle and polar coordinate system.3.Based on the established algorithm model,the circuit module of each part of the algorithm is designed,which mainly includes the design of the external interface,logic function,state control and internal operation circuit of each module.In the design,parallel operation,pipeline,module reuse and other schemes are used to design the operational circuit to ensure resource consumption and timing constraints.4.The system-level design of the test platform is completed around the peripheral control circuits such as the master state machine,algorithm module wrapper and fingerprint sensor of the system.By compiling and synthesizing the design through Quartus,the logical resource consumption and timing analysis report of the each module and overall system can be obtained,and the design can be optimized according to the FPGA devices carried on the test platform.5.Use Modelsim to verify the logic functions of the peripheral device control circuit,each part of the algorithm circuit and the overall authentication system.At the same time,the simulation data is compared with the algorithm model data through MATLAB to complete the functional consistency verification of the system.6.In the test platform of fingerprint authentication system based on Altera EP4CE10F17C8 FPGA,the fingerprint preprocessing and identification matching function are tested.Through the performance evaluation method of fingerprint authentication system,according to the experimental data of the test platform,the recognition rate of the system is 91%,the error acceptance rate is 3%,and the operation time of input mode is about 34.08 ms,and the operation time of matching mode is about34.312 ms.