An Intelligent Real-Time Occupancy Monitoring System with Enhanced Encryption and Privacy

The number of people entering or leaving a building is an essential piece of information that has a lot of practical applications in intelligent building, queue management, and customer service. Vision-based technologies are widely installed in real-time occupancy monitoring systems due to accuracy and reliability. However, monitoring occupancy through unprotected video may disclose privacy of innocent people. Therefore, protecting confidentiality and accurately counting the number of people in real-time scenarios is a severe challenge. Encrypting such videos is one of the promising solutions for maintaining privacy. In this paper, we propose a real-time occupancy monitoring system with Region of Interest (ROI) based light-weight video encryption. People movement is detected through a widely used background model, i.e., Gaussian Mixture Model (GMM) and Kalman filter. Instead of encrypting the whole frame including background, the main idea is to encrypt people present in video via Tangent Delay Ellipse Reflecting Cavity Map System (TD-ERCS). Compared to existing schemes which are mainly based on complete encryption, the proposed method provides partial encryption though cryptographically secure and low-cost computation. The proposed scheme is tested with several different parameters such as correlation, entropy, contrast, energy, Number of Pixel Change Rate (NPCR) NPCR, Unified Average Change Intensity (UACI) and key space. Results from all security parameters have highlighted sufficient security of the proposed scheme.