AI Surveillance Cameras Offer Job Site Security


Construction Technology & Software

Artificial intelligence surveillance cameras can identify people and threats in real time

Workplace shootings and other violent incidents occur all too frequently, and construction job sites, company offices, and shops are not immune.

Kogniz, a tech company in California, has introduced fully artificial intelligence surveillance cameras it says can identify people and threats in real time, using video-based facial recognition and object detection.

“In business environments, surveillance requires identifying problematic people and situations in real time, something humans aren’t good at,” says Daniel Putterman, Co-CEO and head of business for the company. “Kogniz leverages computer vision, artificial intelligence, and machine learning to identify people and threats in real-time with extraordinary accuracy.”

Similar systems are already in use in China, where the government is using a vast array of cameras and facial recognition technology to nab (and shame) criminals.

The Kogniz product, called AICam, is based on the “Kogniz Edge Platform,” a combination of advanced camera processing and cloud services. The architecture allows cameras to be added on-demand without any additional hardware or infrastructure, with multiple cameras working in unison to provide cross-location people identification and recognition.

Key features include:

  • Visual Watchlists. These instantly identify people of concern, such as a known criminal or ex-employee. Using highly accurate facial recognition, Kogniz says it can alert the business when that person is identified.
  • Threat Detection. The system is designed to detect potential threats, immediately warning when it recognizes a weapon or dangerous object.
  • Access Control. The system can watch over a secure area and notify users when an unauthorized person comes into view.
  • Suspect Watch. The company says the system identifies and remembers suspicious behaviors, allowing it to track visit patterns, dwell-time, and other attributes that can help to identify patterns of suspicious behavior.
  • Smart Recording. It also stores relevant video in a private and secure cloud vault that can be accessed from anywhere.

It’s not too hard to imagine such technology also reducing theft, and perhaps even helping to predict theft—such as if a person frequently returns to “case” a particular area.

Kogniz says its products are already in pilot programs with various law enforcement and government agencies, banks, and corporations.