This podcast episode explores the paradox of building modern smart cities: how to achieve total connectivity and public safety without turning security infrastructure into a massive energy drain. The discussion centers on how AI-powered edge processing, exemplified by the Ganz AI Box, solves the false-positive problem in video surveillance while using drastically less power than traditional server farms, and how it can retrofit existing camera networks to avoid e-waste. Finally, the conversation highlights how the CORTROL VMS acts as a city's "central nervous system," unifying sensors and cameras without resource-heavy re-encoding, enabling automated, sustainable building management that supports both security and environmental compliance.

Key Topics

• A smart city uses data to improve infrastructure, reduce environmental footprint, and boost public safety.
• Traditional motion-based surveillance creates excessive false positives, wasting human review time and physical resources like gas and electricity.
• Reducing false positives with AI also cuts unnecessary patrol and guard dispatches, lowering the carbon footprint.
• The Ganz AI Box is a low-power edge device that processes video locally, using only 16 watts (versus ~1000 watts for traditional servers) while handling up to 16 channels.
• The AI Box can retrofit existing "dumb" cameras without replacing hardware, preventing e-waste and saving on labor costs.
• CORTROL VMS integrates data from diverse sensors (cameras, air quality monitors, temperature gauges) without re-encoding video, preserving CPU efficiency and enabling it to run on lightweight devices like PC sticks.
• CORTROL's Event and Action Manager automates responses (e.g., turning off AC in empty rooms), helping buildings meet sustainability regulations like NYC's Local Law 97.
• The overall system shows that security and environmental responsibility can be complementary rather than conflicting when engineered thoughtfully.

Global plastic recycling is stuck because current sorting tech—mechanical and visible-light cameras—can only hit 75-95% purity, well short of the 99% purity manufacturers need to reuse plastic at scale. The core problem: many plastics look identical (or, in the case of black plastic, invisible) to standard cameras since color and shape don't reveal chemical makeup.

Shortwave infrared (SWIR) imaging fixes this by detecting molecular vibration signatures unique to each polymer type, effectively reading an invisible barcode that also cuts through dirt and grime. Paired with the right Computar ViSWIR specialized lenses to prevent focus shift across wavelengths, SWIR-based sorting can process up to 1,000 kg/hour at over 99.5% accuracy, positioning it as a critical enabler for circular economy goals and compliance with emerging recycling regulations.

This episode explores the shift from reactive surveillance to proactive intelligence by integrating Video Management Systems (VMS) and License Plate Recognition (LPR). We break down how turning license plates into searchable metadata enables security teams to move beyond manual tape analysis and toward automated, real-time responses such as instant alerts and barrier control 🛡️.

The discussion highlights the role of neural networks in overcoming environmental challenges—such as obstructed or dirty plates—and the importance of breaking down data silos within existing infrastructure. Featuring a look at CORTROL CVA LPR, we examine how high-speed AI tracking and vehicle behavior analysis are redefining safety for modern facilities and urban environments 🚗.