Font Size: a A A

Design Of Non-Contact Multi-Spectral Palm Vein Recognition System

Posted on:2020-05-21Degree:MasterType:Thesis
Country:ChinaCandidate:J J TaoFull Text:PDF
GTID:2428330572494823Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
In the modern society where digital technology is developing rapidly,the security of information has been paid more and more attention,and the identification technology has been updated.Biometric identification is an emerging technology that uses the intrinsic characteristics of the human body for identification.Compared with the traditional information verification method,it has many advantages such as high security,high reliability,convenient carrying,good anti-counterfeiting,anti-theft and so on.Palm vein recognition as a biometric technology with unique advantages is being gradually developed.Firstly,it is clear that the purpose of this paper is to design a set of non-contact multi-spectral palm vein recognition system for the difficulty of collecting light source modules in palm vein recognition.From the medical point of view,the principle of palm vein imaging and the influencing factors of vein imaging were analyzed.The contrast experiments of vein collection under different near-infrared source wavelengths were carried out.According to the experimental results,the quality of the vein images collected under the common illumination of the 760 nm,850nm and940 nm three-band near-infrared light sources is the best.This design provides a new scheme for the selection of the wavelength of the near-infrared source.Then,the non-contact multi-spectral palm vein image acquisition device is realized,and the hardware selection,layout and uniformity of the 760 nm,850nm and940 nm three-band near-infrared light sources are analyzed and designed,and the hardware driving circuit of the three-band light source is realized.At the same time,the hardware module design is performed on the distance sensing module,the infrared camera module and the filter selection of the acquisition instrument.In addition,image segmentation,ROI extraction methods and factors affecting image quality are studied.The image quality judgment method based on gray uniformity and sharpness is used to judge the image quality more accurately.Finally,aiming at the brightness adjustment of multi-spectral near-infrared light source,a multi-spectral adaptive fast adjustment system based on cuckoo search algorithm is proposed.The image quality judgment is used as the feedback link of the algorithm to realize the fast intensity matching of the three light sources,andhigh-quality vein images can be collected for different people with different palm conditions.Through experiments and simulation verification,the non-contact multi-spectral palm vein recognition system designed in this paper can meet the requirements of fast,accurate,safe and stable system.It can collect high-quality vein images of a large number of different users' palms,which greatly reduces the workload of subsequent image processing and enhances the universal performance and recognition rate of the system identification.The system can meet the high security requirements of different occasions and realize the all-round network requirements.Figure [43] Table [5] Reference [62]...
Keywords/Search Tags:palm vein recognition, light source system, cuckoo search algorithm, multi-spectral, adaptive
PDF Full Text Request
Related items