Research On The Image Processing Of Projection Formula Vein Imaging System

Venous puncture is widely used in clinical diagnosis and treatment of basic skills,the presence of intravenous access difficulties increased the difficulty of vein puncture operation,the use of vein recognition principle to achieve real-time subcutaneous vein surface in the human skin projection,can improve the success rate of vein puncture operation,To reduce the occurrence of medical disputes.In this paper,we study the image processing methods in the real-time imaging system of vein blood vessel from the following three aspects :First of all,according to the principle of vein recognition,build vein image acquisition platform,access to the original vein image on the system hardware technology,software technology at different levels of analysis,the initial establishment of system solutions.Then,the image is corrected,and the corrective algorithm is selected according to the requirements of the system.By contrasting the perspective transformation and the bilinear transformation correction algorithm,the vein images can be corrected efficiently and accurately.Finally,study the vein image enhancement algorithm,in the image preprocessing,using histogram algorithm classic to enhance image contrast,the histogram equalization algorithm with strong limitation suits for image pixel value well-distributed.Contrast limited adaptive histogram equalization by the global limit in the local,to adjust the brightness distribution in local,get more detail information,limits the noise amplification problem.The paper analysis the treatment effect of vein image between HE and CLAHE,the CLAHE algorithm is used to enhance the extraction of vein line.Then,further image segmentation,binaryzation processes are employed after the contrast enhancement of the CLAHE algorithm.The classical OTSU and the local adaptive binaryzation base on the image gray level,and ignore the effect of vein direction.This paper studies characteristics of vein one-dimensional gray,binding the characteristics of the concave curve in the mathematical theory.Vein pattern gray direction field threshold segmentation algorithm is proposed.Calculating the convolution of each pixel in the image with the weight matrix,generating vein gray redistribution image.Get the binary image by thresholding the image.The project completed the research on the image processing of projection vein imaging system,and achieved satisfactory experimental results,which laid the foundation for the development of the follow-up vein projector.