A TSVM-Based Minutiae Matching Approach for Fingerprint Verification

Author(s):  
Jia Jia ◽  
Lianhong Cai
2003 ◽  
Vol 24 (9-10) ◽  
pp. 1349-1360 ◽  
Author(s):  
Yuliang He ◽  
Jie Tian ◽  
Xiping Luo ◽  
Tanghui Zhang

Author(s):  
Tahirou Djara ◽  
Marc Kokou Assogba ◽  
Antoine Vianou

Most matching or verification phases of fingerprint systems use minutiae types and orientation angle to find matched minutiae pairs from the input and template fingerprints. Unfortunately, due to some non-linear distortions, like excessive pressure and fingers twisting during enrollment, this process can cause the minutiae features to be distorted from the original. The authors are interested in a fingerprint matching method using contactless images for fingerprint verification. After features extraction, they compute Euclidean distances between template minutiae (bifurcation and ending points) and input image minutiae. They compute then after bifurcation ridges orientation angles and ending point orientations. In the decision stage, they analyze the similarity between templates. The proposed algorithm has been tested on a set of 420 fingerprint images. The verification accuracy is found to be acceptable and the experimental results are promising. Future work will enhance the proposed verification method by a new template protection technique.


2020 ◽  
Vol 2 (10) ◽  
Author(s):  
BAPPA SARKAR ◽  
JOYASSREE SEN ◽  
MD. ATIQUR RAHMAN ◽  
MD. HABIBUR RAHMAN

Author(s):  
Tahirou Djara ◽  
Marc Kokou Assogba ◽  
Antoine Vianou

Most of matching or verification phases of fingerprint systems use minutiae types and orientation angle to find matched minutiae pairs from the input and template fingerprints. Unfortunately, due to some non-linear distortions, like excessive pressure and fingers twisting during enrollment, this process can cause the minutiae features to be distorted from the original. The authors are then interested in a fingerprint matching method using contactless images for fingerprint verification. After features extraction, they compute Euclidean distances between template minutiae (bifurcation and ending points) and input image minutiae. They compute then after bifurcation ridges orientation angles and ending point orientations. In the decision stage, they analyze the similarity between templates. The proposed algorithm has been tested on a set of 420 fingerprint images. The verification accuracy is found to be acceptable and the experimental results are promising.


Sign in / Sign up

Export Citation Format

Share Document