Artificial Intelligence and Communication Technologies
A Finger Vein Pattern Recognition System Using ROI Extraction and Convolution Neural Network
Authors: Dnyaneshwari P.Wagh, Fadewar H.S and Shinde G. N
Publishing Date: 10-03-2023
ISBN: 978-81-955020-5-9
Abstract
Biometric identification technology has advantages over above traditional method, since each person’s biological characteristic is unique and difficult to counterfeit. At present finger vein recognition is an ideal identification technology. The finger vein pattern in extracted from the input database images, the extracted vein patter in treated as the Region of Interest (ROI) image. This image is given as input to the Convolution Neural Network (CNN) for training. The CNN model extracts featured form the input image and trained the model. The trained model can then be used from recognition of new images. Three fingers per person have been used in the experiments and the proposed method produced good results in recognizing the person. The proposed model produced an accuracy of 100 percent during the testing phase.
Keywords
Finger vein recognition, Convolution neural network, Biometrics, Data acquisition, Vessel identification.
Cite as
Dnyaneshwari P.Wagh, Fadewar H.S and Shinde G. N, "A Finger Vein Pattern Recognition System Using ROI Extraction and Convolution Neural Network", In: Saroj Hiranwal and Garima Mathur (eds), Artificial Intelligence and Communication Technologies, SCRS, India, 2023, pp. 833-846. https://doi.org/10.52458/978-81-955020-5-9-78