Biometrics techniques have been widely applied in practical applications such as automatic control, access control, financial management, etc. The success of these applications depends on much on the quality of the image received from the input device. In practice, the imported image is often of poor quality due to objective noise, or the influence of the receiving device. To overcome this limitation, the data-driven approach has been used by researchers to improve performance when solving biometric problems. Machine learning algorithms are one of the most popular methods of data-driven approach, recent studies show that this method has brought many benefits in biometric systems.

Machine learning (ML) algorithms, such as: K-nearest neighbors, deep neural networks, support vector machines, image feature representation, evolutionary computation etc... suggest a new and effective approach to biometrics problems.

As such, the research team focuses on understanding the application of machine learning algorithms to biometric applications. Going forward, propose new algorithmic improvements to enhance the performance of biometric systems in various practical fields.

Some of the topics of interest include

  • Application of machine learning in biometrics
  • Biometric features (fingerprint biometrics, facial biometrics, etc.)
  • Apply artificial neural network technology and fuzzy logic in biometrics.
  • Multi-biometric combination techniques.
  • Tracking and monitoring the subject's face
  • Deep learning models in biometric recognition.
  • Theory and application of multiple biometrics.
  • Integrate biometric data in personal identification.

Topics

  • Machine learning algorithms (ML), Pattern recognition, Biometric applications.

Member

  • Assoc.Prof.Dr. Le Hoang Thai
  • Dr. Van Thien Hoang
  • PhD student. Tran Binh Long
  • PhD student. Tran Son Hai

Typical research project

  • Development of a number of soft computational models for the problem of tracing human face images looking straight ahead (City level), acceptance 2010.
  • Building a model to extract facial image features for object recognition problem (B2012-18-24; National University level), acceptance in 2013.
  • Multi-state human face image feature descriptor for real-time object tracing application (C2014-18-01; National University level), acceptance 2016.

Possible collaborative activities

  • Implement projects on pattern recognition and applications of biometrics.

Links of demo, website, video

References

Published articles (typical)

  • Thai Hoang Le. Applying Artificial Neural Networks for Face Recognition. Advances in Artificial Neural Systems, Volume 2011, Article ID 673016, 16 pages.
  • Thai Hoang Le and Hoang Thien Van. Fingerprint reference point detection for image retrieval based on symmetry and variation. Journal Pattern Recognition, Volume 45, Issue 9, September, pp.3360-3372, Elsevier Science Inc. New York, NY, USA, ISSN:0031-3203, 2012.
  • Thai Hoang Le, Hung Phuoc Truong, Ha Thi Thanh Do and Duc Minh Vo. On approaching 2D-FPCA technique to improve image representation in frequency domain. Proceedings Of The Fourth Symposium On Information And Communication Technology, SoICT 2013, DaNang, Vietnam, December 5-6, 2013, pp. 172-180.
  • Khoa Dang Dang and Thai Hoang Le. Locality oriented feature extraction for small training datasets using non-negative matrix factorization. Vietnam Journal of Computer Science, Vol 1(4), pp. 257-267, Springer, Vietnam, 2014.
  • Hoang Thien Van, Thai Hoang Le and Tien Ba Dinh. Efficient palmprint identification using novel symmetry filter and alignment refinement. International Journal of Biometrics, Volume 7, Issue 3, pp.213-225, Inderscience Publishers (IEL), 2015.
  • Thai Hoang Le. On Approaching Heuristic Weight Mask To Enhance LBP-Based Profile Face Recognition System. Indian Journal of Science and Technology, Volume 9, Issue 17, May 2016.
  • Hai Son Tran, Thai Hoang Le and Thuy Thanh Nguyen. The Degree of Skin Burns Images Recognition using Convolutional Neural Network. Indian Journal of Science and Technology, Volume 9, Issue 45, December 2016.
  • Long Binh Tran and Thai Hoang Le. Multimodal Personal Verification Using Likelihood Ratio for the Match Score Fusion. Computational Intelligence and Neuroscience (CIN), vol.2017, Article ID 9345969, 9 pages, 2017.
  • Khoa Tan Truong and Thai Hoang Le. Video-Based Face Recognition Using Shape And Texture Information In 3D Morphable Model. JP Journal of Heat and Mass Transfer – Pushpa Publishing House, Allahabad, India, 2018.