• A.Vineetha
• G.Sreenivasulu

with the advent of high power, cheap computing and greater complexity, biometric authentication has become possible at every scale because of its safer nature and also user-friendly. According to different researchers, vein biometric is a good biometric trait among others such as fingerprint, palm and finger veins, eyes, voice, signature, gait and DNA for authentication systems. In this paper, PalmDorsal (Back of hand) vein pattern authentication using a Convoluted Neural Network (CNN) is proposed. The main advantage of CNN over other traditional approaches is its ability to simultaneously extract features, reduce dimensionality of data, and classifying it using single network structure. Also, the method requires only minimal steps for image pre-processing since the CNN is robust to noise.This paper proposes a new approach to authenticate the individuals based on the hand dorsal vein images and the knuckle shape features using CNN. The proposed system is a fully automated and it uses a contactless, low-cost near IR imaging device to capture hand vein images. The minutiae points: vein bifurcations and vein endings, are extracted from the hand vein image and along with them knuckle points are used to perform authentication. The matching scores are generated in two stages: (i) hierarchical matching score from the four sets of triplets generated from binarized vein image (ii) the knuckle tip distances and vein map length. The weighted average of the matching scores is used to authenticate an individual. The proposed system produced satisfactory results and provides a more user friendly way of authenticating individuals. In existing system palm vein recognition work has been done. And in proposed method dorsal recognition work has been done. We have to compare accuracy is improved than palm vein recognition. The proposed system developed on MATLAB 2013a version.

Image Processing, Biometrics, Contactless hand based authentication, knuckle shape, hand dorsal veins etc.

"Based On Convolutional Neural Networks Dorsal Hand Vein Recognition for High Security Applications ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 5, page no.c786-c792, May-2022, Available :http://www.jetir.org/papers/JETIR2205402.pdf

"Based On Convolutional Neural Networks Dorsal Hand Vein Recognition for High Security Applications ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 5, page no. ppc786-c792, May-2022, Available at : http://www.jetir.org/papers/JETIR2205402.pdf