Duos Technologies Group, Inc., through its operating subsidiary Duos Technologies, Inc., has been awarded a U.S. patent, Use of Artificial Intelligence to Detect Defects in Trains and Method to Use. The company describes the patent covering “supervised machine learning that proactively detects defects and identifies derailment risks, setting a new benchmark for rail industry safety enhancement.”
Duos said the patent “reinforces our strong commitment to improving rail safety through technology. Utilizing the Duos proprietary AI process significantly enhances the effectiveness of mechanical railcar inspections to aid in derailment prevention, further improving safety, and streamlining operations.”
“The timely delivery of actionable intelligence is a key objective of Duos inspection technology,” said Duos CTO Jeff Neccia. “This patent covers technology that seamlessly integrates supervised Machine Learning into the inspection process and introduces the HITL (human in the loop) feature, which empowers rail industry professionals to conduct focused, accurate and efficient inspections. I believe the patent was awarded due to our ability to perform real-time analysis of high-resolution image data and metadata, making it ideal for identifying defects, preventing hazards, and ensuring safety.”
“Our AI-driven approach is a game-changer for the industry,” said Duos CEO Chuck Ferry. “By combining our deep understanding of the rail sector with AI technology, we’re not only helping prevent derailments, but setting a new standard for safety in the rail industry through innovation and strategic advancements in artificial intelligence.”
Duos noted it AI teams “are staffed with rail industry experts, leading to a high performance factor for automated detections while avoiding questionable results generated by theoretical models developed in a laboratory without specific rail industry expertise.”
Duos was recently awarded a multi-year contract from a Class I railroad using this AI technology. The company did not disclose the customer.