Critical Infrastructure (CI) can be considered an asset, a facility, equipment, a network or a system, or a part of an asset, a facility, equipment, a network or a system, which is necessary for the provision of an essential service (EC Directive 2022/5557). Any disruption or destruction of the CI’s operation would have a significant impact on the vital societal functions of health, security, and economic or social well-being of people, and would cause cascading effects on other interdependent systems resulting in catastrophic results and long-term harmful consequences. On the other hand, the penetration of IoT and AI achievements in physical assets have transformed the nature of CIs, by alleviating the boundaries between the physical and cyber worlds. Hence, the Cyber-Physical Systems (CPS), which are the core of the current CI systems, are capable of the digital control of physical processes. However, the operation of CIs in this complex environment generates a wide spectrum of challenges and risks related to physical/natural threats, human-made and cyber malicious activities. Reducing the vulnerabilities and improving resilience of CIs have become a priority for the authorities around the globe to enhance the CIs protection and their operational continuity.
In this framework, M4D, exploiting the gained knowledge and experience from its participation in several projects in the Security domain, is aiming to develop tools enable to strengthen the surveillance capabilities of CIs and to increase the situational awareness of CIs operators. The analysis, with innovative deep-learning algorithms, of multimedia content information captured by CCTVs and UAVs, can lead to the timely detection of suspicious objects, recognition of malicious activities and unauthorized persons for their entrance in CIs premises. These outcomes can be indexed and fused semantically in order to assist in the estimation of the severity level of a potential attack to CIs premises, serving the needs for real-time monitoring and mitigating the risk.