Page 32 - Security Today, November/December 2023

Unlocking Deeper Insights for Safer Environments By Vishwa Vijoyendra Narayan Data is the lifeblood of effective physical security measures. The ability to gather, store and analyze video surveillance data is essential for safeguard- ing people and places. However, simply collecting the data is not enough. To leverage the vast amount of video surveillance data most organizations are generating for things like crowd control and reaction times to security concerns, organizations can use the power of Artificial Intelligence (AI) to gain valuable insights into their video surveillance data for better decision making and to deliver better customer experiences. As a result, they gain stronger insights to help them make bet- ter decisions and deliver improved experiences. THE PHYSICAL SECURITY DATA DILEMMA Surveillance cameras, access control systems, intrusion detection sensors, and cybersecurity tools, just to name a few, generate an immense volume of data. This data is mostly unstructured which could be anything from text documents and images to audio and video files. Unstructured data makes up about 80% of all data gathered and is expected to continue its exponential rise. This sheer volume and diversity of unstructured data presents both opportunities and challenges. Historically, analyzing such data has been a resource-intensive task, limited by human capabilities and time constraints. How- ever, AI helps organizations cut down on the manual intensive process in a way that’s accessible and effective. This technology brings with it the power of automation and advanced analytics, enabling the enrichment of video surveillance data in ways that were previously unattainable. Here are some key aspects of how AI is transforming the physical security landscape. Real-time Analysis. Algorithms powered by AI can analyze video surveillance data provided by objects like cameras or audio devices in real-time, allowing for an immediate or automated re- sponse to potential threats. For instance, surveillance footage can be monitored for unusu- al activities, like loitering or movement in restricted areas, with alerts triggered as soon as anomalies are detected. This analysis also provides live information on factors such as crowd move- ment and density, while also offering guests information to help navigate a space easily and efficiently, ultimately creating a posi- tive experience. Pattern Recognition. AI is trained on previously collected data to identify patterns and trends that might escape the human eye. Within these patterns, AI can detect recurrent activities that may signify a security breach or vulnerability, and it can even predict future security threats based on historical data. Behavioral Analysis. This advanced technology can learn typi- cal behavior patterns and flag deviations. For instance, in a corpo- rate setting, AI can recognize when an employee accesses restricted areas at unusual times, potentially indicating insider threats. Multi-Modal Data Integration. Physical security data comes from multiple sources. AI can integrate data from multiple sourc- es, such as video feeds, access logs, and sensor data, to provide a holistic view of security. This integrated approach offers a com- prehensive perspective that is invaluable. Scalability. AI technology works 24/7 without fatigue, ensur- ing continuous monitoring and analysis. This frees up physical security and surveillance teams to focus on more critical projects. THE STORAGE IMPERATIVE With the rise of AI and analytics, organizations can provide more holistic physical security than ever before. However, these capabilities depend on having enough computing power, storage, and easy access to their data to unlock AI’s potential. In addition, growing complexity of hosting such varied and powerful security applications requires a scalable storage system with overarching management abilities. Previous solutions like NVRs weren’t built to scale easily and cost-effectively to keep up with the massive amount of data being generated today. When it comes to physical security data, organizations should seek out solutions that provide simple integration and manage- ment capabilities and are software-defined so there’s no risk of losing video data if hardware fails. A software solution that uni- fies previously stranded server and storage resources can scale and provide better quality of service control to allow both high and low performance workloads to co-exist, and to create better data accessibility. Analytics impact every piece of data, so it is crucial that it all remains easy to access and manage. A critical component of the data management process is the ability to archive data for ex- tended periods of time in a cost-effective way. A storage solution should also make it easy to access archived data so that users can use it for analysis and gain insights to drive business outcomes. A SECURE FUTURE PAVED WITH AI AI is revolutionizing the way we approach data. By enriching stored video data with AI-driven insights, we can achieve a level of security and analysis that was once beyond our reach. The ability to analyze vast amounts of data in real-time, recognize patterns, and respond to anomalies is invaluable in our quest to protect people, assets, and information. As AI technology continues to evolve and storage solutions adapt right alongside it, the future of physical security for any organization looks brighter, safer, and more adaptive than ever before. Vishwa Vijoyendra Narayan is the director of product management at Quantum. 32 NOVEMBER/DECEMBER 2023 | SECURITY TODAY ARTIFICIAL INTELLIGENCE