Research On Near Infrared Imaging Of Human Body Vein

Human vein imaging technology has become a hotspot on biometrics and medical device research,especially the computer technology,the continuous development of medical image and pattern recognition and other disciplines,the human body vein has been as an important biological characteristics used in various fields.This technology successfully resolved the traditional pressure pulse,postoperative repair need clear vein texture problems,could help doctors accurately rapid intravenous injection,check veins postoperative recovery and other issues.Firstly,the principle of human vein imaging are analyzed in detail and in-depth analysis,through the selection of the various parts of the device in hardware circuit,including FPGA minimum system board,CMOS image sensor and infrared LED light source module,display module,and completed the design and implementation of human vein image acquisition platform;Secondly,the experiment proves that the system has a strong real-time performance,which shows that the veins of human veins are clear and have good connectivity and good stability.Because the vein image collected by the human vein imaging device is inevitably polluted by all kinds of noises in the process of transmission,transformation and processing,it is difficult to distinguish the quality of the image through the observation of the human eye.Therefore,at last,we evaluated the quality of human vein images processed and processed by corresponding algorithms,and laid a solid foundation for selecting clear images.The main contents are as follows:(1).CLAHE enhancement and NiBlack algorithm are given to the human vein images collected under different wavelengths of near-infrared light,and then the image quality was compared before and after treatment by multiple sets of experiments.The experimental results show that in the near-infrared reflection light,940nm monochromatic light can collect the best quality arm blood vessels image,760nm and 940nm mixed light compared to the second,760 nm monochromatic light collection effect is not ideal.Therefore,mixed light is selected as light source in this experiment;(2).Because the dorsal hand compared to some parts of the human body has a richer vein pattern,then we selected the dorsal vein images for this study.The collected vein images were contrast enhanced,Gaussian emphasis filtering,vein segmentation and refinement and other treatment,each step of the treatment are selected through the comparative experiment the best processing algorithm for the next step,the vein pattern after treatment is clearly visible,Visual effect is better,which provides a solid theoretical basis for biometric identification,postoperative venous recovery;(3.).The quality of human vein image acquisition platform,image transmission and image processing may cause various kinds of noise pollution,resulting in image sharpness impaired.Therefore,the image quality evaluation method is used to evaluate the articulation of human vein images.The evaluation result is consistent with the subjective evaluation results of human,which proves that the quality of vein images collected in this paper is better;(4).Through the comparative experiment,the performance of the collecting device was tested,and the experimental method with the best effect was selected.The experimental results show that the vein images collected are clear and the connectivity is good,which lays the foundation for further research basis.