Development Of Dual Beam Vein Imaging System

Human hand vein imaging technology is widely used in venipuncture and other medical fields.With the development of computer technology,it has become one of the new research focuses in the field of biometrics.Commonly used single transmission type and reflection type near infrared region have different effects in different palm palm vein imaging technology and make the vein imaging effect is low,cannot effectively in clinical use and difficult to satisfy the requirement of feature recognition technology,in order to solve this situation,A system of dual-beam imaging combining reflection and transmission is designed to obtain the vein image of the hand.Based on the physical structure of human hand and the interaction of light and biological tissue characteristics,according to the principle of near-infrared imaging,innovatively proposed a design scheme based on the combination of reflection and transmission imaging,the technical idea is to obtain the reflected and transmitted palmar vein images respectively,and then the two vein images were denoised and enhanced respectively,finally,image fusion was performed to obtain a high quality hand vein angiogram with excellent clarity and contrast.The system design is divided into two major parts of hardware and software.Firstly,the hardware system is composed of an illumination system and an image collector,that is,reflective and transmissive light sources and their corresponding hardware control circuits and CCD cameras.Among them,STM32 is selected as the main control chip of the circuit,the wavelength of the near-infrared LED of the reverse transmission light source is 810 nm and 850 nm respectively,and the transmission light source is an adaptive light source based on the near infrared receiver module.Secondly,the software system is on the VScode software platform,using Python language based on Open CV and PIL and other libraries for algorithm processing.Algorithms mainly include denoising based on the combination of DCT domain and median filteringmorphology filtering,skin color recognition based on YCb Cr space,image enhancement based on local mean-standard deviation and CLAHE algorithm,image registration fusion based on Surf algorithm and GLCM-based quality assessment.Finally,the software interface is designed by Py Qt5.Based on the principle of near-infrared imaging of hand veins,this paper designs and implements a dual-beam hand vein imaging system with complete functions,a combination of reflection and transmission,and is of great value in clinical identification of venipuncture and biometric feature recognition.