Related Experiment Video
Updated: May 26, 2026
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
Effective fingerprint quality estimation for diverse capture sensors
Shan Juan Xie1, Sook Yoon, Jinwook Shin
1Department of Electronics and Information Engineering, Chonbuk National University, 664-141 Ga Deokjin-Dong, Jeonju, Jeonbuk 561-756, Korea. shanj_x@jbnu.ac.kr
Sensors (Basel, Switzerland)
|December 14, 2011
Summary
This study introduces a versatile fingerprint quality estimation system that adapts to various sensors. It improves recognition accuracy by effectively filtering low-quality images, enhancing overall system performance.
Area of Science:
- Biometrics
- Computer Vision
- Machine Learning
Background:
- Fingerprint recognition system performance relies heavily on input image quality.
- Assessing fingerprint quality requires sensor-specific features, posing adaptation challenges.
Purpose of the Study:
- To develop an adaptive fingerprint quality estimation system for diverse capture sensors.
- To enhance the accuracy and reliability of fingerprint recognition systems.
Main Methods:
- A novel system combining orientation certainty, local orientation quality, and consistency features was designed.
- The system extracts basic and next-level features applicable across various capture sensors.
- A Support Vector Machine (SVM) classifier was employed for image quality assessment.
Main Results:
- The proposed method demonstrated superior accuracy compared to existing techniques.
- The system effectively eliminated residue images from optical and capacitive sensors.
- Coarse images from thermal sensors were successfully filtered out.
Conclusions:
- The developed quality estimation system offers improved accuracy and adaptability for fingerprint recognition.
- This approach enhances the robustness of biometric systems across different sensor types.
