Finger-vein recognition system /
Biometric feature-based recognition systems are trending on the data security business. The unique features of the human body are used in security systems also known as a biometric system, such as voice, fingerprint, iris recognition, face recognition, etc. Among all, finger-vein recognition is at t...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Kuala Lumpur :
Kulliyyah of Engineering, International Islamic University Malaysia,
2019
|
| Subjects: | |
| Online Access: | Click here to view 1st 24 pages of the thesis. Members can view fulltext at the specified PCs in the library. |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Biometric feature-based recognition systems are trending on the data security business. The unique features of the human body are used in security systems also known as a biometric system, such as voice, fingerprint, iris recognition, face recognition, etc. Among all, finger-vein recognition is at the peak of its popularity amidst the consumers, as well as security researchers and organizations are willing to use this technology commercially. However, finger-vein recognition is exclusive because of the veins which are underneath of human's skin rather than outside structure. Thus, compared to other biometric technology finger-vein is different plus it is almost impossible to fabricate, even two twins' finger-veins are fully different from each other. To establish a finger-vein recognition system, it required a huge amount of finger-vein data and for that reason, a finger-vein acquisition device is made as well as created the dataset. To create the finger-vein acquisition device, near infrared (NIR) imaging technology had been used and built a dataset which is split into two datasets, one for training and another for testing. To train the dataset convolutional neural network (CNN) is used via transfer learning of Alexnet using MATLAB in order to finding out the recognition accuracy. After that, a real-time finger-vein recognition system is developed which captured finger-vein images of individuals and after running those images through the recognition system generates a recognition rate. This research obtained 100% accuracy when tested with the stored dataset, however, in real-time 99%, predictive accuracy is received in several experiments which is notably a satisfactory result. |
|---|---|
| Physical Description: | xiv, 72 leaves : illustrations ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 60-63). |