DOI 10.1007/s11063-006-9004-y Neural Processing Letters (2006) 23:305–323 © Springer 2006 Face Matching in Large Database by Self-Organizing Maps TOMMY W. S. CHOW ⋆ and M. K. M. RAHMAN Department of Electronic Engineering, City University of Hong Kong, Kowloon, Hong Kong. e-mail: eetchow@cityu.edu.hk Abstract. A novel self-organizing map (SOM) based retrieval system is proposed for performing face matching in large database. The proposed system provides a small subset of faces that are most similar to a given query face, from which user can easily verify the matched images. The architecture of the proposed system consists of two major parts. First, the system provides a generalized integration of multiple feature-sets using multiple self-organizing maps. Multiple feature-sets are obtained from different feature extraction methods like Gabor filter, Local Autocorrelation Coefficients, etc. In this platform, multiple facial features are integrated to form a compressed feature vector without concerning scal- ing and length of individual feature set. Second, an SOM is trained to organize all the face images in a database through using the compressed feature vector. Using the organized map, similar faces to a query can be efficiently identified. Furthermore, the system includes a rel- evance feedback to enhance the face retrieval performance. The proposed method is compu- tationally efficient. Comparative results show that the proposed approach is promising for identifying face in a given large image database. Key words. face matching, feature integration, self-organizing map, relevance feedback 1. Introduction Face recognition has been an active research topic since the last two decades in the areas of biometrics, pattern recognition, and computer vision. Face recognition has a wide variety of applications on commercial, security, and law enforcement. These applications can briefly be categorized into two types, namely face verifica- tion and face identification. Face verification is simply used to verify if a given image belongs to a specific person, whilst face identification is a more compli- cated problem. For an unknown given face image, face identification matches with all face images in a database. This is quite a demanding task from the perspec- tive of pattern recognition. Although there has been a rapid growth of large-scale databases, researchers have focused only on the accuracy with small databases. The issues of scalability and computational speed, which are important to large data- base application, have, however, been overlooked. In large database face recogni- tion problem, response time, search, and retrieval efficiency have been crucial in addition to recognition accuracy. In this work, we consider face identification as ⋆ Corresponding author.