Smart roads to the rescue: New AI-based system aims to prevent roadkill
The South Korean government is pioneering efforts to reduce roadkill with the deployment of advanced artificial intelligence technology.
On Wednesday, the Ministry of Environment launched a pilot program for its Animal Road Accident (Roadkill) Prevention System, designed to detect wildlife on roadways using image recognition AI. The system is currently being tested on two provincial roads: one in Yangpyeong, Gyeonggi Province, and the other in Pyeongchang, Gangwon Province.
This innovative system combines AI-enabled CCTV cameras and LiDAR (light detection and ranging) sensors to monitor continuously for animals on the roads. When wildlife is detected, a warning message -- 'Wild animal detected within 150 m ahead' -- is displayed on LED boards situated 150 meters before the hazard zone.
The ministry said in a press release that by incorporating deep-learning algorithms, the technology analyzes patterns in wildlife behavior, including species types and their activity times. This predictive capability allows for timely alerts, helping drivers to reduce their speed and remain vigilant.
'Drivers will be able to recognize animals ahead, slow down and take precautionary measures, which is expected to significantly reduce accidents,' the ministry said.
The project was initiated in March 2025 after a memorandum of understanding was signed in July 2024 among the Ministry of Environment, Posco DX, the Korea National Park Service (KNPS) and the National Institute of Ecology (NIE).
One provincial road segment in Yangpyeong and another in Pyeongchang, both recognized as hotspots for animal-related incidents, were chosen for this initial rollout. The ministry led the project, providing administrative support, while Posco DX was responsible for developing and installing the system. KNPS and NIE played vital roles in identifying suitable locations and assessing the system's effectiveness.
Looking ahead, the ministry plans to extend the system's reach to three additional high-risk areas, including Hoengseong in Gangwon Province and Namwon in North Jeolla Province, by 2027.
Officials believe the system will not only help prevent roadkill but also reduce traffic accidents and vehicle damage, enhancing both driver safety and biodiversity conservation.
To celebrate the launch, the ministry held a completion ceremony at the Cheongun Township Office Welfare Center in Yangpyeong, attended by approximately 50 participants. The event featured a live demonstration of the AI system in action.
'This initiative signifies a collaborative approach to addressing wildlife road accidents,' said Kim Tae-oh, director of the Nature Conservation Bureau at the Ministry of Environment. 'We are committed to pursuing projects and policies that foster corporate involvement in biodiversity conservation.'
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Korea Herald
6 hours ago
- Korea Herald
Tving and Wavve merger gets green light with conditions
Amid Netflix's dominance in the local streaming market, merger of South Korea's popular streaming platforms approved on condition of fee freeze through end of 2026 Popular South Korean streaming services Tving and Wavve have been given the green light for a merger. On Tuesday, the Fair Trade Commission granted conditional approval for the merger of the two platforms, imposing a measure that requires the companies to maintain their current subscription fee levels through the end of 2026 in order to prevent potential price hikes tied to the launch of bundled products. Entertainment giant CJ ENM currently owns Tving, while Wavve operates as a joint venture between terrestrial broadcasters KBS, MBC and SBS, and telecom operator SKT. In its review of the deal, the antitrust regulator said it examined whether subscription fees for consumers could rise post-merger. According to the FTC, both Tving and Wavve command a substantial base of loyal subscribers. Due to the availability of exclusive content, these subscribers tend to exhibit relatively low price sensitivity. In particular, the FTC said subscribers who highly value exclusive content, such as Tving and Wavve's live broadcast channels and KBO League baseball coverage, would find it difficult to switch to competing services, even if prices were to increase under a bundled model. To address concerns over potential fee increases and other anticompetitive effects in the streaming market, the FTC said that it approved the merger on the condition that Tving and Wavve maintain their existing pricing plans from the date the corrective order is issued — Tuesday — through Dec. 31, 2026. Furthermore, should Tving and Wavve be integrated into a single service during this period, the FTC said the new service must introduce a new pricing plan that is comparable to their current plans in both price range and service offerings, and maintain this new structure through the same 2026 deadline. As of January 2025, Tving and Wavve rank second and fourth in the market, respectively, accounting for 16.5 percent and 9 percent of total streaming service usage time on smartphones in South Korea. According to the same data, Netflix accounts for 61.1 percent. Tving and Wavve have been pursuing a merger since late 2023 as part of efforts to expand their combined market share.
Korea Herald
6 hours ago
- Korea Herald
AI teaching assistant shows real promise at S. Korea tech university KAIST
South Korea's top science and engineering university has rolled out a custom-built AI teaching assistant to help students in large graduate courses, and early results show it helped reduce repetitive student questions while encouraging more active, self-directed learning. At Korea Advanced Institute of Science and Technology, a team of researchers led by AI graduate school professor Choi Yoon-jae and design professor Hong Hwa-jung developed a virtual teaching assistant, or VTA, that can answer student questions at any time, with responses tailored to specific lectures and coursework. The tool was piloted last fall in a programming class for AI, taken by 477 graduate students. It's the first time such a system has been tested at scale in a Korean university setting. What sets the KAIST VTA apart is that it isn't a generic chatbot. Instead, it runs on what's called a retrieval-augmented generation model, which pulls directly from course materials like slides, coding exercises and lecture videos. When students ask a question, the system finds the most relevant content and formulates a response based on that context. This means answers are grounded in what's actually taught, not just generated from a general AI model. Over 50 percent of students used the system regularly during the 14-week semester, generating nearly 3,870 questions and answers. Students without a strong background in AI or coding were among the most active users, suggesting the VTA helped close knowledge gaps for those new to the subject. These figures come from internal usage data collected by KAIST during the semester. The system didn't just benefit students. According to lead TA and doctoral researcher Kwon Soon-jun, it reduced the number of routine questions from students, such as basic concept definitions or explanations already given in class. That allowed human teaching assistants to focus more on deeper, more complex issues. Compared to the previous year's course, the volume of questions requiring direct responses from TAs dropped by around 40 percent, based on data compiled by Professor Choi. Students also appeared more comfortable asking questions through the VTA than to human TAs, especially when it came to theoretical topics. Surveys conducted by Choi's research team before, during and after the course showed that students became increasingly confident in the system's reliability, and those who had been hesitant to speak up in class reported higher satisfaction levels when using the AI assistant. The VTA's source code has been released publicly on GitHub to encourage adoption by other educators and researchers. The work was also accepted to the Industry Track at ACL 2025, one of the leading international conferences in natural language processing
Korea Herald
12 hours ago
- Korea Herald
Cognex Introduces OneVision: A Breakthrough Cloud Platform for AI-Powered Machine Vision
New Training Environment Delivers Scalable Automation and Solves Key Barriers to Industrial AI Adoption NATICK, Mass., June 10, 2025 /PRNewswire/ -- Cognex Corporation (NASDAQ: CGNX), the global technology leader in industrial machine vision, today announced the upcoming full launch of OneVision ™, a cloud-based platform that will transform how manufacturers build, train, and scale AI-powered vision manufacturers build, train, and scale AI-powered vision applications. "OneVision is the result of nearly a decade of industrial AI innovation at Cognex. By combining our advanced AI technology with excellent customer experience, we continue to unlock new possibilities for our customers," said Carl Gerst, Executive Vice President of Vision and ID Products at Cognex. "We're simplifying the entire application development process; helping our customers move faster, scale more efficiently, and maximize the value of their automation investments." Designed for the demands of global manufacturing, OneVision tackles the most common barriers to wide-scale deployment of AI-powered machine vision: With the OneVision software platform, Cognex expands its industry-leading portfolio of machine vision products, unlocking new applications, accelerating AI adoption, and allowing users to solve complex vision tasks faster, more reliably, and at scale. OneVision is currently available to select customers on the In-Sight® 3800 and 8900 vision systems and will be released on additional Cognex products in early 2026. and will be released on additional Cognex products in early 2026. About Cognex Corporation For over 40 years, Cognex has been making advanced machine vision easy, paving the way for manufacturing and distribution companies to get faster, smarter, and more efficient through automation. Innovative technology in our vision sensors and systems solves critical manufacturing and distribution challenges, providing unparalleled performance for industries from automotive to consumer electronics to packaged goods. Cognex makes these tools more capable and easier to deploy thanks to a longstanding focus on AI, helping factories and warehouses improve quality and maximize efficiency without needing highly technical expertise. We are headquartered near Boston, USA, with locations in over 30 countries and more than 30,000 customers worldwide. Learn more at
