
This database is acquired using a FBI-compliant slap fingerprint sensor that are widely deployed at the border crossings and used in various national ID programs, including Aadhaar. This imaging system is actually an addon system which just placed a low-cost imaging camera on the top of fingerprint sensor to simultaneously acquire slap finger-knuckle images in a single-fingerprint-imaging-shot and without any additional imaging system. This is the first such system and developed at the Hong Kong Polytechnic University. Above GUI interface illustrates simultaneously acquired finger-knuckle and fingerprint images using 4-4-2 imaging and also respective knuckle images after the segmentation.
The system captures the 4-4-2 slap fingerprint images and corresponding knuckle
images simultaneously. It uses a commercial FS64 slap fingerprint sensor and a
5 mega-pixel camera placed at the top of such sensor. There are a total 116
subjects participated in such imaging. For each subject, 5 imaging instances
are acquired. The imaging was done under ambient illumination in indoor and
outdoor environment. The images are labelled using the following convention:
File naming convention: AAA-B-CC-D.bmp
|
Naming |
Description |
Values |
|
AAA |
Subject ID |
Ranges from 001 to 116 |
|
B |
Slap/Knuckle Image |
1: Slap Fingerprint Image;
2: Finger Knuckle Image |
|
CC |
Finger/Knuckle Type |
Ranges from 01 to 21 |
|
Fingerprints/Major Knuckles |
01: LL, 02: LR, 03: LM, 04:
LR, 05: LT, 06: RT, 07: RI, 08: RM, 09: RR, 10: RL |
|
|
Slap images |
11: Left-4 Fingers, 12:
Two-thumbs, 13: Right-4 Fingers |
|
|
Major Knuckles |
14: LLa,
15: LRa, 16: LMa, 17: LRa, 18: RIa, 19: RMa, 20: RRa, 21: RLa
|
|
|
D |
Instance |
Ranges from 1 to 5 |
|
Others |
LL-Leftlittle,
LR-LeftRing, LM-LeftMiddle,
LI-LeftIndex, LT-LeftThumb,
RL-RightLittle, RR-RightRing,
RM-RightMiddle, RI-RightIndex,
RT-RightThumb
|
The entire database as detailed above is made available to support further
research and development efforts. The below sample finger dorsal images
acquired along with slap fingerprint images. The first row from the same user
using 4-4-2 imaging while the second row illustrates other users’ images
acquired under the ambient imaging. These images also illustrate the knuckle
region contours and respective bounding boxes in green color.




The images in the below figure illustrate simultaneously acquired fingerprint images from a slap fingerprint sensor, corresponding to a use with knuckle images in the first image of the above figure.
The database is being made available for the researchers from June 2023 onwards. Interested researchers should follow following steps to acquire "The Hong Kong Polytechnic University Single Shot Slap Fingerprint and Finger Knuckle Images Database"
All the rights of the The Hong Kong Polytechnic University Single Shot Slap Fingerprint and Finger Kncukle Images Database are reserved and commercial use/distribution of this database is strictly prohibited. All the technical reports and papers that report experimental results from this database should provide due acknowledgement and reference. The reference should appear as 'The Hong Kong Polytechnic University Single Shot Slap Fingerprint and Finger Knuckle Images Database'. Questions regarding this database can be directed to Ajay.Kumar@polyu.edu.hk.
[1] Zhenyu Zhou and Ajay Kumar, " Finger-Knuckle Assisted Slap Fingerprint Identification System for Higher Security and Convenience.” IEEE Transactions on Information Forensics and Security, vol. 19, pp. 441-454, 2024.
Back to Databases
Page