Related Experiment Video
Updated: Aug 4, 2025
11:34
High-resolution, High-speed, Three-dimensional Video Imaging with Digital Fringe Projection Techniques
Published on: December 3, 2013
15.7K
Monocular 3D Fingerprint Reconstruction and Unwarping
Summary
Contactless fingerprint acquisition offers hygiene benefits but faces challenges from perspective distortion. A new learning-based method reconstructs 3-D finger shapes, improving contactless fingerprint recognition accuracy.
Area of Science:
- Biometrics
- Computer Vision
- Image Processing
Background:
- Contactless fingerprint acquisition offers advantages over traditional methods, including improved hygiene and reduced skin distortion.
- Perspective distortion in contactless fingerprint images degrades recognition accuracy by altering ridge frequency and minutiae locations.
Purpose of the Study:
- To develop a novel method for reconstructing 3-D finger shapes from single contactless images.
- To mitigate perspective distortion in contactless fingerprint images for enhanced recognition.
Main Methods:
- A learning-based shape-from-texture algorithm was employed to reconstruct 3-D finger geometry.
- The reconstructed 3-D shape was used to unwarp raw fingerprint images, correcting for perspective distortion.
Main Results:
- The proposed method demonstrated high accuracy in 3-D reconstruction of finger shapes from contactless fingerprint databases.
- Experimental matching results showed significant improvements in accuracy for both contactless-to-contactless and contactless-to-contact-based fingerprint comparisons.
Conclusions:
- The developed learning-based approach effectively addresses perspective distortion in contactless fingerprint recognition.
- This technique enhances the accuracy and reliability of biometric identification systems utilizing contactless fingerprint data.
