Smart Public Safety Video Surveillance System

Abhishek Djeachandrane is a research scientist at Airbus Defence and Space’s AI Connectivity Lab, France. His research interests include AI, data science, computer networks, QoE and trustworthy systems.Said Hoceini is Associate Professor and Head of the N&T Department at IUT CV.UPEC, France. His research focuses on routing algorithms, QoS/QoE, and bioinspired artificial intelligence approaches.Serge Delmas is an engineer at Airbus Defence and Space, Secure Land Communications, France. He leads the Research & Technology Projects and Innovations team, driving cutting-edge solutions to enhance future emergency services.Abdelhamid Mellouk is Full-time University Professor, Director of the IT4H High School Engineering Department and Head of the TincNET Research Team, UPEC, France. He is also the founder of Network Control Research and Curricula activities at UPEC, President of the Policies and Programs commission at the National Council for Scientific Research and Technologies, a HCERES Expert, a CNU member and Co-President of the DS-AI Systematic Deep Tech Hub.

• Preface ixList of Acronyms xiIntroduction xviiChapter 1. Literature Review on an End-to-End Video Surveillance System for Public Safety 11.1. General description: human threats in urban areas and abnormal situation detection 21.2. Analytics for video surveillance 21.2.1. Crowd behavior analysis 41.2.2. Traffic analysis 91.2.3. Environment analysis 111.2.4. Individual behavior analysis 121.2.5. General human threat-centric urban situation analysis 141.3. System architecture for video surveillance 211.3.1. Network architecture 211.3.2. Computing infrastructure 231.4. Analytics and architecture: studies and reflections 241.4.1. Threats: from cyber-to-physical space or physical-to-cyber space? 251.4.2. Video quality impact on video surveillance: from monitoring to task-specific analytics 271.4.3. End-to-end measurement: from traditional QoE to task-specific QoE 281.5. Challenges 291.6. Conclusion 31Chapter 2. A Development Platform for Integration and Testing 332.1. Introduction 332.2. Proposed framework - QoE-driven SA-centric DSS 342.2.1. High-level view of the system: reinforcement signal and QoE 342.2.2. Detailed system framework: SA-centric DSS 362.3. Use case -Airbus DSSLC's target market 482.3.1. Introduction 482.3.2.Challenges 482.3.3. Purposes 492.3.4. Application case: Airbus DS SLC's business opportunity 492.3.5. Target system: Airbus DS SLC's flagship product 532.4. Conclusion 54Chapter 3. A Multi-Criteria Enriched Corrective Signal with Endogenous, Exogenous and Human Factors 553.1. Context 553.2. Problem statement 573.3. Proposals 583.3.1. QoP for endogenous factor assessment 613.3.2. Task-specific QoE for endogenous, exogenous and human factors 663.4. Conclusion 74Chapter 4. A Situational Awareness-centric Predictive System for Anomaly Detection 774.1. Context 774.2. Baseline 784.3. Problem statement 794.4. Proposals 814.4.1. Feature extraction experimentation and reviewing 824.4.2. Capability-oriented classifier study 834.4.3. Result-oriented classifier study 1014.5. Conclusion 117Chapter 5. Towards an Autonomic Intelligent Video Surveillance System 1195.1. Context 1195.2. Problem statement 1205.3. Proposals 1215.3.1.Time-based control 1225.3.2. Event-triggered control 1245.4. Conclusion 142Conclusions and Perspectives 143References 149Index 161