Unlocking the Future of Cameras: AI-Powered Video Analytics at the Edge

In the fourth episode of the IoT Pulse podcast, we take an in-depth look into the innovative world of AI and video analytics with Hisham Wakim from BstSecurity.
Listen on:
Spotify Spotify Youtube Youtube Apple Music Apple Music
Guests
Hicham Wakim
BstSecurity
Hachim Wakim is the Founder and CEO of BSTSecurity, a leading provider of innovative security solutions. With over 15 years of experience in the security industry, Hachim is a seasoned expert in developing and implementing advanced security strategies tailored to the unique needs of businesses. Under his leadership, BSTSecurity has become known for its customer-centric approach, integrating cutting-edge technology with personalized service to deliver exceptional security outcomes. Passionate about leveraging technology to create safer environments, Hachim is committed to driving continuous innovation and growth in the security landscape.
Co-Hosts
Chris & Kirill
EyeOTmonitor
Chris and Kirill are the co-hosts of the IoT Pulse podcast show. Chris has a proven track record in building successful teams and fostering strategic partnerships that push the boundaries of what's possible in tech. Kirill is known for his insightful perspectives and hands-on approach, Kirill has been instrumental in shaping the cybersecurity landscape, making complex security challenges more manageable for businesses across various sectors.
 

Introducing BstSecurity: A New Approach to Video Surveillance

BstSecurity is not just another camera company; they are at the forefront of creating tailored solutions that redefine the video surveillance industry. Founded in 2012 in Montreal, BstSecurity, under the leadership of Hisham Wakim, has been focused on delivering customized cameras, software, and AI capabilities to integrators, distributors, and project developers.

Hisham explained how BstSecurity started with basic IP cameras but quickly moved towards more advanced video analytics. Their early work in perimeter protection involved linking fixed cameras to pan-tilt-zoom (PTZ) cameras using specialized algorithms to track vehicles and people while minimizing false alarms. This innovative approach highlights BST’s commitment to solving specific industry problems rather than offering generic camera solutions.

Beyond Traditional Video Analytics: The Power of AI at the Edge

While many companies view video analytics and AI as interchangeable, BstSecurity takes a more nuanced approach. According to Hisham, traditional video analytics often involve simple rules-based systems, like line-crossing detection or motion detection, that can identify objects or movements. However, AI-based video analytics go further by learning the environment and distinguishing between various objects, such as a human versus an animal, or differentiating between car colors.

BstSecurity leverages machine learning models directly on edge devices—whether on cameras or AI boxes—to enhance real-time processing and minimize network load. This approach is critical in environments like city surveillance or traffic management, where constant real-time analysis is required without transmitting vast amounts of video data over the network. Instead, metadata is sent, reducing network strain and allowing for more scalable and efficient operations.

Key Benefits of Edge Processing in Video Analytics

The episode delved into the critical benefits of processing video data at the edge:

  1. Reduced Latency and Increased Efficiency: By pushing processing to the edge, BstSecurity significantly reduces latency, making real-time analytics possible without overwhelming network bandwidth. This approach allows for quicker decision-making in critical scenarios, such as detecting accidents in city traffic or managing crowd control in public spaces.
  2. Enhanced Privacy and GDPR Compliance: Edge processing also helps maintain privacy by keeping sensitive data on-site. BST’s approach to face recognition involves generating unique IDs for individuals without storing actual images, ensuring full GDPR compliance and privacy protection.
  3. Cost-Effectiveness and Resilience: By distributing processing power across multiple edge devices, BstSecurity creates a more resilient system. If a single device fails, it affects only a tiny part of the overall network. This decentralized approach also reduces the need for high-powered centralized servers, cutting costs and improving system reliability.

Real-World Applications: From Retail to Smart Cities

Hisham shared several compelling use cases that illustrate the practical applications of BstSecurity’s technology:

  • Retail Optimization: For retail clients, BST provides solutions that can accurately count people and analyze demographics without recording or storing video footage. This information helps retailers target their marketing more effectively and make informed decisions about store layouts, optimizing customer experience and sales.
  • City Surveillance and Traffic Management: In city surveillance, BST’s solutions can predict traffic jams before they happen or detect accidents in real time, sending only relevant metadata to central servers. This reduces network congestion and allows city authorities to respond quickly to incidents.
  • Industrial and EV Garage Monitoring: BST also leverages thermal cameras for industrial applications, such as monitoring transformers for overheating or managing electric vehicle (EV) garages to prevent fires. The AI-driven thermal cameras can detect overheating issues that traditional systems might miss, providing a proactive approach to safety in high-risk environments.

The Future of Video Analytics at the Edge

Looking ahead, BstSecurity is pushing the boundaries of what video analytics can achieve. The company is developing solutions for various sectors, including free zones where container tracking and automated gate control can streamline logistics and reduce congestion. AI-based security and operational management systems promise to transform multiple industries by minimizing operational costs, enhancing security, and providing better management tools.

Hisham also highlighted the importance of managing these distributed edge devices effectively. As the edge becomes a critical part of infrastructure—akin to network routers or switches—the ability to monitor, update, and optimize these devices is vital. This is where solutions like IoT Monitor come into play, providing centralized management of edge devices to ensure high availability and optimal performance.

Conclusion: A Verticalized Future for Video Analytics

As AI and machine learning technologies continue to advance, the future of video analytics is moving toward more specialized, verticalized solutions tailored to specific industries. BstSecurity is leading this charge by developing more innovative, more efficient, and more privacy-conscious surveillance solutions that secure environments and enhance operational effectiveness. With innovative products like their small, powerful thermal cameras and edge-based AI boxes, BST is set to redefine how businesses think about video analytics and surveillance.

As Chris Nixon and Kirill wrapped up the episode, it was clear that the future of video analytics lies at the intersection of AI, edge computing, and smart infrastructure. BstSecurity’s forward-thinking approach offers a glimpse into a more connected, secure, and efficient world where every camera, sensor, and AI box is crucial in building smarter cities and safer communities.

Stay tuned to IoT Pulse for more insights on how IoT is transforming industries and reshaping the future!