Research On Finger Capacitance-induced Biometric Recognition Method Based On System Correlation Identification

In today’s society,multimodal biometrics has become a trend,and the research on the biometrics of the hand can make the hand multimodal biometrics get a better recognition rate.In the early stage,the research team proposed a biometric identification method based on palm bioelectrical impedance spectrum(BIS)measurement and a biometric recognition method based on finger-induced capacitive reactance spectrum(CRS)measurement.The former has a low recognition rate.Although the latter can obtain better results,the data calculation is complicated.Based on the previous research,this paper proposes a finger-capacitance sensing biometric recognition method based on system-related identification,which aims to reduce the complexity of system data calculation and improve the recognition rate of hand biometrics.1.The pseudo-random property of m-sequence is analyzed,and the principle of system correlation identification is studied.When the m-sequence is used as the excitation signal,the time-domain impulse response of the system can be obtained by finding the cross-correlation function of the input and output signals,thereby reflecting the system behavior.In this paper,system behavior refers to finger capacitance sensing characteristics.The relevant parameters of the m-sequence selected for this subject are discussed in detail.2.Designed a finger capacitance induction biometric measuring instrument with FPGA as the core architecture.The m-sequence signal generator was successfully implemented and verified by Verilog hardware programming language.A bipolar voltage source with a stable conversion effect is constructed to convert the unipolar voltage generated by the FPGA into a bipolar voltage.The data acquisition and storage system of ADC+FPGA+SD card is designed.The experimental platform can quickly complete signal generation,acquisition,storage and processing3.The factors affecting the acquisition effect of the finger capacitance sensor are analyzed,and the finger capacitance sensing instrument designed by the research is used to collect,analyze and compare the sensors of different design methods and sizes.Finally,a finger capacitance induction sensor with better performance is selected.4.Preliminary data acquisition and classification recognition experiments were performed using the designed finger capacitance sensing measurement system.Data collection was conducted on 50 volunteers,and a training sample library and a test sample library were established.BP neural network algorithm and SVM algorithm are used to classify and recognize biological features.The recognition accuracy of BP neural network algorithm can reach 76%,and the recognition accuracy of SVM algorithm can reach 73%.This paper makes full use of the pseudo-random characteristics of m-sequence signals and explores a new method of biometric recognition based on the principle of system correlation identification.The research results preliminarily prove that this method has certain feasibility and provides a new idea for multi-modal recognition of hand biometric features.