Person Vein Identification using CNN || Deep Learning || Venkat Projects || Hyderabad
TO PURCHASE OUR PROJECTS IN ONLINE (OR) OFFLINE
CONTACT:VENKAT PROJECTS
NAME: VENKATARAO GANIPISETTY
Mobile & WhatsApp :+91 9966499110
Email :[email protected]
Website:www.venkatjavaprojects.com
About Project:
In this project we are employing Convolution Neural Network (CNN) to identify person based on finger vein. CNN will extract features from images automatically and then used this features for person identification and does not require any hand crafted features like finger minutiae or palm area HAAR cascaded files which contains finger and palm information and this files designed by humans and may contains error which result into inaccurate identification.
All existing biometric algorithms are dependent on humans hand crafted features whose prediction accuracy is not up to the mark. In propose work we are employing machine learning SVM algorithm and deep learning CNN algorithm and then evaluating both performance in terms of accuracy and confusion matrix.
To train both algorithms we are using 30 different finger veins images dataset downloaded from KAGGLE website and below is the dataset URL. All this images are captured using Thermal cameras.