Robust Face Detection in Airports
- PDF / 1,337,383 Bytes
- 7 Pages / 600 x 792 pts Page_size
- 60 Downloads / 205 Views
Robust Face Detection in Airports Jimmy Liu Jiang School of Computing, National University of Singapore, Science Drive 2, Singapore 117559 Email: [email protected]
Kia-Fock Loe School of Computing, National University of Singapore, Science Drive 2, Singapore 117559 Email: [email protected]
Hong Jiang Zhang Microsoft Research Asia, Beijing Sigma Center, Beijing 100080, China Email: [email protected] Received 25 December 2002; Revised 3 October 2003 Robust face detection in complex airport environment is a challenging task. The complexity in such detection systems stems from the variances in image background, view, illumination, articulation, and facial expression. This paper presents the S-AdaBoost, a new variant of AdaBoost developed for the face detection system for airport operators (FDAO). In face detection application, the contribution of the S-AdaBoost algorithm lies in its use of AdaBoost’s distribution weight as a dividing tool to split up the input face space into inlier and outlier face spaces and its use of dedicated classifiers to handle the inliers and outliers in their corresponding spaces. The results of the dedicated classifiers are then nonlinearly combined. Compared with the leading face detection approaches using both the data obtained from the complex airport environment and some popular face database repositories, FDAO’s experimental results clearly show its effectiveness in handling real complex environment in airports. Keywords and phrases: S-AdaBoost, face detection, divide and conquer, inlier, outlier.
1.
INTRODUCTION
A human face detection [1, 2, 3] system can be used for video surveillance and identity detection. Various approaches, based on feature abstraction and statistical analysis, have been proposed. Among them, Rowley and Kanade’s neural network approach [4], Viola’s asymmetric AdaBoost cascading approach [1], and support vector machine (SVM) approach [5] are a few of the leading ones. In the real world, the complex environment associated with the face pattern detection often makes the detection very complicated. Boosting is a method used to enhance the performance of the weak learners (classifiers). The first provable polynomialtime boosting model [6] was developed from the probably approximately correct (PAC) theory [7], followed by the AdaBoost model [8], which has been developed into one of the simplest yet effective boosting algorithms in recent years. In pattern detection and classification scenarios, the training input patterns are resampled in AdaBoost after every round of iteration. Easy patterns in the training set are assigned lower distribution weights; whereas the difficult patterns, which are often misclassified, are given higher distribution weights. After certain rounds of iteration, based on
the values of the distribution weights assigned to the training input patterns, input training patterns can be classified into inliers (easy patterns) and outliers (difficult patterns). When AdaBoost is used to handle scenarios in complex environment with many outlie
Data Loading...
