comes even more powerful, Berndt said. AI-driven video systems are transforming school bus safety by turning cameras into active safety tools rather than passive recorders, noted Jetha. “Instead of relying on footage that must be manually reviewed after an incident, AI enables continuous monitoring of what’s happening around the bus and inside the vehicle, identifying when something isn’t right and taking timely action,” she added. AI-based identification improves school bus safety by
combining speed, consistency and prevention, Jetha said. “Like in traditional human review, these technologies
have had certain preset parameters that you could create thresholds within and use the older systems to create some sort of easier way to look for a quick video that you had in question,” said Bryer. “A human had to program the thresholds. They were customized based on what the end user’s parameters they wanted to be notified about with AI. “You can look at specific data points that you are having to set and it has got more of a generalized understanding of certain situations like rolling a stop sign, analyzing data as far as counting passengers on and off the vehicle and
what kind of stops have more danger zones,” he contin- ued. “It is more like live capability. Now we can use AI to prevent things from happening. AI can provide an alert.” As AI technology advances, camera systems are
increasingly able to flag altercations or risky behavior by analyzing human movement and interactions, and identifying patterns that deviate from normal activity, Jetha noted. “With more powerful processors and AI-enabled cameras, this analysis can now occur on the vehicle in real time, while also supporting cloud-based review after the fact, helping districts prioritize safety-critical events more quickly and consistently than manual video review alone,” she said. Jetha said Safe Fleet is conducting internal
proof-of-concept work using next-generation cameras with embedded AI chips capable of running advanced models directly on the bus. “One area of focus analyzes motion patterns,” she said.
“AI can identify behavior that appears unusual or po- tentially unsafe such as physical altercations or sudden incidents without relying solely on raw video review. “We’re also building on existing AI work such as
www.stnonline.com 45
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
