Research On Infrared Vein Image Processing Algorithm Based On Vein Features

It is very important to obtain accurate,clear and intuitive vein information,whether it is to assist intravenous injection by accurately locating human vein vessels,or to extract vein features for identity authentication.However,the quality of the infrared vein image captured by the infrared camera is low.The vein information extracted by the existing infrared vein image processing algorithms often has a large number of breakpoints,and the thickness of each sub-segment of the vein is uneven,which is greatly different from the actual vein.Therefore,in view of the problem that the vein information extracted from the infrared vein image is not accurate and intuitive,this paper proposes the processing algorithm of the infrared vein image from three aspects which are the enhancement and segmentation of the vein image,the repair of breakpoints and the repair of the uniformity of the thickness based on the features of the vein.For the enhancement and segmentation of the infrared vein image,this paper first designs the pre-processing algorithm for the low contrast,uneven illumination and noise problems of the infrared vein image.Then,on the basis of the pre-processed vein image,the continuity,uniformity of thickness and the consistency of overall vein direction are summarized.After that,this paper analyzes the gray distribution curve of vein images,and proposes a vein segmentation algorithm base on vein direction.Through the fusion of the segmentation results of four directional operators in different directions,vein information can be extracted as completely as possible from the vein images.As for the repair of vein breakpoints in the vein information extracted by segmentation algorithm,this paper first designs a noise removal algorithm to remove a large number of point noises in the vein binary image,providing a basis for subsequent repair of the breakpoints.Then,based on the overall direction consistency of vein image,this paper designs a multi-scale vein breakpoint repair algorithm.After that,combined with the overall vein direction extracted from the vein direction image obtained during vein segmentation,the breakpoints of vein information can be repaired and the continuity of veins can be improved.In terms of the repair of the uniformity of vein thickness,this paper firstly divides the vein into sub-segments according to the bifurcation structure of the vein,and analyzes the necessity of the differently repair of the vein sub-segments thickness.Then,this paper models the sub-segment veins,proposes the estimation algorithm of the sub-segment veins thickness,and makes a targeted thickness repair for each sub-segment veins to improve the uniformity of the thickness of the veins.In this paper,the proposed infrared vein image processing algorithm based on vein features is experimented respectively,and the effect of image processing is evaluated from the perspective of subjective and objective evaluation.The experiment shows that the vein segmentation algorithm based on vein direction,the multi-scale vein breakpoint repair algorithm based on overall vein direction and the vein thickness uniformity repair algorithm proposed in this paper can extract accurate,clear and intuitive vein information from the infrared vein images of different directions.