Contactless Finger Knuckle Identification using Smartphones

 

This project is attempting to develop a smartphone based online system to automatically identify a person by using their finger knuckle image. The key objective is to exploit user-friendly biometric which may have least privacy concern to enhance security of the data in smartphone. Several studies on the fingerprint identification have suggested that a large proportion of user population, i.e. elderly, manual laborers, etc. do not have fingerprints of sufficient quality to be used in automated fingerprint identification system.  Recent large scale proof of concept study conducted by UIDAI has also estimated  that ~1.9% of subjects cannot be reliably authenticated using their fingerprints. Therefore some alternative imprints like finger knuckle patterns can be exploited for authentication and requires further research efforts.

      The final product from this project is the finger knuckle authentication smartphone application which can operate from smartphone with Android Platform with environment version 2.3.3. This project will develop some specialized algorithms for the finger knuckle detection, image pre-processing and region segmentation. The automatically detected and segmented finger knuckle images will be used to encode finger knuckle pattern phase information using log-Gabor filters. The initial implementation of various modules in this online application is performed in C/C++ programming language with OpenCV library. This project also develops a user-friendly graphical user interface for users to enroll and authenticate themselves. The developed system can therefore acquire finger knuckle image from the smartphone camera and automatically authenticate the genuine users. This project has also developed a new smartphone based finger knuckle database from 561 finger knuckle images (from 187 clients) in real imaging environment and this database is now made available for other researchers and developers. The video clip in the following demonstrates the performance and usage of the first version of this smartphone application.

Download - The Hong Kong Polytechnic University Mobile Phone Finger Knuckle Database, 2012

Reference: Kam Yuen Cheng and Ajay Kumar, "Contactless finger knuckle identification using smartphones," Proc. IEEE Intl. Conf. BIOSIG 2012, pp. 1-6, Darmstadt, Germany, September 2012