Superficial vein imaging technology in the medical field can provide medical vein image,auxiliary venipuncture,improve the success rate,with strong applicability of the technology in the field of identification also plays an important role,so shallow vein imaging technology has been widespread concern.Traditional shallow vein imaging technology does not provide in-depth information on veins,but for venipuncture or identification,the lack of deep vein information may lead to failure of the venipuncture,unable to identification problems.Aiming at the hand vein image can not provide in-depth information,we present a binocular vision technology for hand vein three-dimensional reconstruction method.In this paper,superficial vein imaging binocular vision technology for application background for binocular vision 3D reconstruction of the platform,effective camera calibration,preprocessing and vein of the 3D image four main issues were studied using Matlab,Visual Studio 2013 software has been verified.First,the human hand vein as the acquisition target,designed and built a three-dimensional reconstruction of binocular vision platform.Depending on the environment selected camera model,design near-infrared light to determine the spatial position of each device.Secondly,the use of Matlab completed two cameras effective calibration.Analysis of the existing types of calibration plate calibration is proposed based on the electronic board electronic device screen,compared to traditional calibration plate under the same conditions,to improve the calibration accuracy.Again,complete pretreatment on hand vein image.According to hand vein in the near-infrared light imaging characteristics,completed the de-noising,enhancement,and adaptive thresholding and connected domain threshold,completed the vein image extraction within the lines.Finally,on the back of the hand vein completed a three-dimensional reconstruction.On the left and right cameras capture images SAD match the calculated disparity map two images overlap region calculated overlap region pixels based on parallax depth data to veins texture image index,screened shot dorsal vein of the three-dimensional image,and completed the back of the hand compared with the dorsal vein three-dimensional image of the two-dimensional image of a vein. |