Image Acquisition System Based On Fingerprint And Finger Vein

With the rapid development of information technology,information security issues have been highly valued by people.Traditional identity authentication methods which are difficult to carry,easy to be forged and of low security are gradually replaced by new biometric authentication.Compared with the current mature fingerprint recognition technology,finger vein recognition technology as a new generation of biometric recognition technology has many advantages such as living certification,non-traceable,user-friendly.In the biometric recognition system based on fingerprint or finger vein,the quality of images is directly related to the overall performance of the identification system.So we do the following research focusing on fingerprint and finger vein image acquisition and image quality enhancement:1.Several biometric recognition technologies and characteristics in the current mainstream are introduced,and the development and research status of image acquisition,image quality enhancement,image quality assessment and market application about the finger vein recognition technology are described.2.In this thesis,we design a fingerprint and finger vein image acquisition system platform,which consists of hardware system and software system:(1)The hardware system is divided into three modules: The first is the core control module base on STM32F407ZGT6 chip,which is comprised of power supply circuit,serial port circuit,debugging interface circuit,real-time clock interface circuit.The second is a finger vein image sensor device with adjustable light source and controllable image,which includes an external structure,a near-infrared light source,a OV2640 image sensor and so on;and the third is a fingerprint image acquisition device based on the FPC1011F3 sensor.The hardware platform design process and implementation principles are descriped in great detail,including the processor and peripheral chip selection,image acquisition device,the overall structure and design requirements,the working principles and driver design of the sensors.The designed hardware system has the advantages of lower cost,stronger expansion,higher real-time and higher quality image.(2)The software system achieves two functions: one is the fingerprint and finger vein image acquisition,which realizes the fingerprint and vein image acquisition and file operation functions by using the USB cable to communicate with the hardware system;another function is the vein image quality enhancement.The software system can improve the quality of original vein images through the region of interest,normalization,filtering and enhancement of finger vein image.The test results show that the software system has theadvantages of simple interface,high control ability,fast image acquisition speed and obvious image enhancement effect.3.In this thesis,a new method of finger vein image quality classification based on hybrid SVM is proposed.The algorithm is divided into three steps: firstly,the one-dimensional and two-dimensional eigenvalues of the finger vein image space are calculated as the feature vector of image quality,and then the quality of the enhanced image is classified to reduce the subjective error of the original image;Finally,a large number of venous image samples are trained by using hybrid SVM with better generalization performance.Our experiments verify the feasibility and effectiveness of the algorithm and show that the hybrid SVM classifier has better learning and generalization ability for finger vein image quality classification.