With the increasing public concerns of security and personal data privacy worldwide, security and privacy become an important research area. This research area is very broad and covers many application domains. The security and privacy aware computing research group at HKBU focuses on (1) privacy-preserved computing, (2) video surveillance, and (3) secure biometric system.

Privacy-preserved Computing
Concerns on the data privacy have been increasing worldwide. For example, Apple was reportedly fined by South Korea’s telecommunications regulator for allegedly collecting and storing private location data of iPhone users. Indeed, the privacy concerns raised by both end-users and government authorities have been hindering the deployment of many valuable IT services, such as data mining and analysis, data outsourcing, and mobile location-aware computing. Thus, in response to the growing necessity of protecting data privacy, our research group has been focusing on developing innovative solutions towards robust information services --- to support these services while preserving users’ personal privacy.

Video Surveillance
With the growing installation of surveillance video cameras in both private and public areas, the closed-circuit TV (CCTV) has been evolved from a single camera system to a multiple camera system; and has recently been further extended to a large-scale network of cameras. One of the objectives of a camera network is to monitor and understand security issues in the area under surveillance. While the camera network hardware is generally well-designed and roundly installed, the development of intelligent video analysis software lags far behind. As such, our group has been focusing on developing video surveillance algorithms such as face tracking, person re-identification, human action recognition. Our goal is to develop an intelligent video surveillance system.

Secure Biometric System
With the growing use of biometrics, there is a rising concern about the security and privacy of the biometric data. Recent studies show that simple attacks on a biometric system, such as hill climbing, are able to recover the raw biometric data from stolen biometric template. Moreover, the attacker may be able to make use of the stolen face template to access the system or cross-match across databases. Our group has been working on face template protection, multimodality template protection, and biometric de-duplication. The goal is to develop a secure biometric system.

Related Projects