Resolution and Distortion Compensation based on Sensor Evaluation for Interoperable Fingerprint Recognition

The resolution difference and image distortion due to sensor characteristics lead to considerable difficulties when matching fingerprints between disparate sensors. The purpose of this study is to compensate the image resolution and distortion before fingerprint features are extracted in order to increase the fingerprint recognition rate between disparate sensor systems. This paper presents a compensation algorithm, both in template level and in image level. In the template level compensation, the unit vector method is proposed for compensation of minutiae direction. The distortion can be compensated by utilizing the horizontal and vertical resolutions of each fingerprint sensor. The proposed algorithm has been tested over Korean Fingerprint Recognition Interoperability Alliance's database constructed by nine commercial sensors, and demonstrates very promising results.