The Hong Kong Polytechnic University 3D Fingerprint Images Database

Traditional 2D fingerprint identification requires subjects to press their fingers against some hard surface, like glass, silicon or polymer. Such imaging and often results in partial or degraded quality fingerprint images. Such frequent degradation in fingerprint image quality is often attributed to skin deformations, moisture, reside of finger dirt/sweat/slips, and smear or due to sensor noise. Contactless 2D fingerprint systems can provide hygienic solutions to such problems and can cope-up with the residue of previous fingerprint impressions which can also be a potential security threat. In addition to these benefits, contactless 3D fingerprint systems can offer significantly improved accuracy and advanced anti-spoofing capabilities than those can be availed from contactless 2D fingerprint systems.

    Lack of any publicly available 3D fingerprint database is one of the key problems to promote research and development efforts in contactless 3D fingerprint identification. Therefore we have acquired 3D fingerprint database from 260 (240 [1]+ 20 [2]) clients, six images per clients, and the entire database of 1560 3D fingerprint images is being made available publicly to promote further research efforts in this area. As detailed in reference [1]-[2],  database was acquired during February 2012 to May 2012 in our university and uses 10920 2D fingerprint images to reconstruct entire 3D fingerprint database.  

 

Download and Copyright

The database is made available freely for the researchers from July 2013 onwards. Interested researchers should follow following steps to acquire the 'The Hong Kong Polytechnic University 3D Fingerprint Images Database'.

     1. Visit the online application link and provide all the details in this application form;

     2. The request confirmation will be sent by email and the applicants can download the database using instructions in the email;

     3. You can also download the source images, calibration data (pixel orientations and normalization images) and a sample/typical code to help in reconstructing 3D fingerprints from this weblink. The password for extracting this additional data is provided in the readme file (accessible after you have successfully downloaded 3D fingerprint database using online application link).

 All the rights of The Hong Kong Polytechnic University 3D Fingerprint 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 with citation to references [1]-[2]. All questions regarding this database can be directed to Ajay.Kumar@polyu.edu.hk.

 

References

[1] Ajay Kumar, Cyril Kwong, "Towards contactless, low-cost, and accurate 3D fingerprint identification," Proc. CVPR 2013, pp.3438-3443, Portland, Oregon, June 25-27, 2013

[2] Ajay Kumar, Cyril Kwong, "Towards contactless, low-cost, and accurate 3D fingerprint identification," IEEE Transactions on Pattern Analysis and Machine Intelligence,  vol. 37, no. 3, pp. 681-696, March 2015

Back to Databases Page