This page was machine-translated and may differ from the original. View original
Daejeon Regional Office of Land and Transport & Kolon Industries Cases Revealed: Generative AI Video Analysis in Traffic and Industrial Safety Sectors
Nota announced on the 1st that it has applied NVIDIA VSS (Video Search and Summarization) to its video surveillance solution NVA (Nota Vision Agent) and introduced it to the Daejeon Regional Office of Land and Transport's traffic control system and Kolon Industries' Gimcheon Plant 2. As an NVIDIA Connect Partner, Nota applies video AI technology to its solutions to configure site-specific functions.
NVA is a generative AI video surveillance solution based on Vision Language Models (VLM) that analyzes situations within video and provides search, summarization, and reporting functions. NVIDIA VSS is a video AI technology that supports the implementation of search, summarization, and question-and-answer functions in video data.
In the field of traffic control, NVA was applied to the road CCTV control system operated by the Daejeon Regional Office of Land and Transportation. The system detects unexpected situations such as accidents, fires, and obstacles, and automatically organizes traffic information and response scenarios by lane. According to Nota, the system achieved 99% accuracy in the Ministry of Land, Infrastructure and Transport's ITS basic performance evaluation.
In the field of industrial safety, NVA was implemented at Kolon Industries' Gimcheon Plant 2 in collaboration with Kolon Benit. On-site, it is being utilized for verifying worker safety, monitoring hazardous areas, and identifying potential safety violations. Instead of directly reviewing video content, operators can search for necessary situations or check summarized information.
Nota explained that it is sharing operational experience gained during field application with the NVIDIA VSS team. In the future, it plans to expand the scope of application to industrial safety, public safety, traffic control, and smart cities by incorporating the multi-agent architecture presented in VSS 3.1 into NVA.
Generative AI-based video surveillance is being utilized to reduce the burden of manual verification in existing CCTV monitoring and to rapidly provide response information for accidents or dangerous situations. However, expanding its application in the field is expected to require accuracy verification, optimization for specific operating environments, and a robust personal information protection system.
본 기사에 대한 정정·반론·추후보도 청구는 보도 청구 안내를, 그간 게재된 보도문은 정정·반론보도 모아보기를 참고해 주세요.














