In this episode, we're exploring the rising security threats in retail—like organized crime and workplace violence—and why stores are shifting from reactive to proactive defense. We dive into the AI Box, a new tool that connects to existing cameras to add deep-learning-powered predictive analytics.

Key takeaways:

• Why traditional tools like standard IP cameras and EAS tags aren't enough anymore.
• How the AI box uses real-time object classification to slash false alarms and improve response times.
• What the Retail Worker Safety Act means for large retailers and their violence prevention programs.

In this episode, we cover why telecentric lenses are so important for manufacturing. Standard lenses distort size with distance, but telecentric lenses maintain the same magnification. We talk about why this true-to-scale accuracy is critical for things like reading text on curved surfaces or inspecting semiconductor wafers. We also look at Computar's new LensConnect series, which adds remote focus via I2C or USB so production lines don't have to stop for manual lens adjustments.

Episode highlights:

• How telecentric lenses fix perspective distortion to give a perfectly accurate view.
• Real-world uses, like reading optical characters on curved surfaces and measuring minute details on wafers.
• How Computar’s LensConnect series uses remote focus to speed up automated inspections without manual tweaking.
• What happens when you combine these lenses with AI, and how it shifts the role of human inspectors in quality control

Security systems aren’t just “record and review” anymore. In this episode, we unpack how Ganz CORTROL positions the VMS as a building’s central nervous system—fusing video, access control, AI identity analytics, mobile, and IoT/audio sensors into real-time workflows that can act (not just alert). We’ll get into what matters to practitioners: enterprise scale, resilience, evidentiary integrity, and the operational line between actionable intelligence and false-positive fatigue.

This episode explores AI-driven machine vision and how it's transforming logistics and warehousing. We explain how the deep-learning brains of machine vision (CNNs) need equally precise eyes: lens choice is crucial. Expect a practical look at how vision systems boost throughput, safety, and uptime—and the obstacles that stand in the way.

Key points

• Machine vision turns pixels into decisions (routing, damage detection, OCR).
• Scan tunnels and high‑speed cameras for automation sorting and quality control.
• What an integrated vision-enabled safety (forklift ADAS, PPE compliance) reduces accidents.
• Why the choice of machine vision lens is crucial and how geometric distortion, aberration, and poor edge resolution undermine measurement, OCR, and routing at scale.

You can't fix bad optics with good code — geometric distortion and poor edge resolution break measurement, OCR, and routing at scale. 🎯👀
Accurate machine vision drives throughput and uptime. 🎧

Imagine security that sees what's coming before it happens. 🔮

In our latest episode, "Threat Detection in Real-Time: The Rise of Predictive Security," we're exploring how AI-powered edge computing is shifting from reactive surveillance to predictive intelligence.

No cloud delays. No privacy compromises. Just instant threat detection with the Ganz AI Box Pro, powered by machine learning.

This episode examines why lens selection is a decisive factor in UAV imaging performance, often outweighing gains from higher-megapixel sensors or advanced software. We'll discuss:

• Optical tradeoffs—coverage versus detail via field of view, resolution limits governed by MTF and sensor pixel pitch, and the importance of edge sharpness and low distortion for photogrammetry, mapping, and measurement accuracy.
• UAV-specific operational constraints, including motion and vibration that demand fast apertures for higher shutter speeds, and environmental stresses (shock, sustained vibration, temperature swings, dust/moisture) that can cause focus drift and inconsistent data.
• Optical instability to downstream failures in AI systems through dataset drift and reduced model confidence.
• Ruggedized lens case study (Computar MPZ‑R series) and a practical checklist for matching lenses to mission geometry, sensor format, shutter-speed needs, distortion tolerance, resolving power, and durability requirements.

For decades, capturing both visible and shortwave infrared required two separate cameras. This episode explores how breakthrough sensor technology finally unified both spectrums in a single chip—and why it's transforming industries.

• The Problem — Traditional silicon sensors can't detect infrared, forcing expensive dual-camera systems
• The Solution — Copper-to-copper bonding enables Sony's IMX992 & IMX993 sensors to capture full visible-to-SWIR spectrum (400-1700nm) on one chip
• Real-World Impact — Revolutionizing semiconductor inspection, plastic recycling, food quality control, and medical imaging
• Cost Revolution — Simplified design dramatically reduces system complexity and price
• Future Applications — Opens doors to medical diagnostics and long-range imaging previously impossible with traditional tech

Learn how to upgrade your legacy security systems into smart surveillance networks without costly hardware replacements. This episode discusses how a non-invasive AI unit can seamlessly integrate with your existing cameras to provide predictive behavioral analysis, compliance monitoring, and privacy-sensitive threat detection. With the Ganz AI Box, you can transform the costs associated with security infrastructure from capital expenditures into a straightforward configuration.

• Painless AI Integration: Add AI intelligence to legacy security systems without replacing existing hardware
• Predictive Threat Detection: Move beyond object detection to intent recognition through advanced behavioral analysis
• Real-World ROI: Monitor workplace safety, compliance violations, and unauthorized access across hundreds of camera feeds
• Privacy-First Design: Dynamic face and license plate masking protects bystander anonymity while preserving security intelligence