Skip to main content

Multiclient Identification System Using Adaptive Probabilistic Model


This paper aims at integrating detection and identification of human faces in a more practical and real-time face recognition system. The proposed face detection system is based on the cascade Adaboost method to improve the precision and robustness toward unstable surrounding lightings. Our Adaboost method innovates to adjust the environmental lighting conditions by histogram lighting normalization and to accurately locate the face regions by a region-based-clustering process as well. We also address on the problem of multi-scale faces in this paper by using 12 different scales of searching windows and 5 different orientations for each client in pursuit of the multi-view independent face identification. There are majorly two methodological parts in our face identification system, including PCA (principal component analysis) facial feature extraction and adaptive probabilistic model (APM). The structure of our implemented APM with a weighted combination of simple probabilistic functions constructs the likelihood functions by the probabilistic constraint in the similarity measures. In addition, our proposed method can online add a new client and update the information of registered clients due to the constructed APM. The experimental results eventually show the superior performance of our proposed system for both offline and real-time online testing.

Publisher note

To access the full article, please see PDF.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Yu-Wen Shou.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License ( ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Lin, CT., Siana, L., Shou, YW. et al. Multiclient Identification System Using Adaptive Probabilistic Model. EURASIP J. Adv. Signal Process. 2010, 983581 (2010).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: