Search across all content
projectThe system was developed to detect various banner types, including political party, private, and public banners, with high accuracy.
Illegal banners present social challenges, including the deterioration of urban aesthetics and risks to public safety. Existing passive management approaches have limited effectiveness and demand significant administrative resources. This project seeks to harness AI technology to optimize banner management and rapidly resolve community concerns.
This project established an AI-based banner detection system and verified its efficiency through field validation.
First, a banner dataset was constructed using public data and self-collected data for AI training. This secured high-quality data reflecting diverse environments and conditions.
An AI-based banner detection model was then developed, and the model's performance in detecting and classifying banner locations was optimized. The system was developed to detect various banner types, including political party, private, and public banners, with high accuracy.
For real-time banner detection system validation, field testing was conducted using a vehicle- mounted 4-channel CCTV system, verifying system performance in Busan's actual urban environment.
Furthermore, a dashboard-based platform was built to manage banner detection results and statistical data. This supports efficient banner management and rapid administrative processing.
This system enables immediate response to illegal banner issues, which is expected to improve urban aesthetics and public safety. It will also reduce administrative resource consumption and enable efficient management, contributing to decreased social and economic costs. Introduction of AI-based banner detection technology is expected to reduce urban maintenance personnel and costs, and to decrease illegal banner installations, etc.
Launch year: 2024
Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.1, https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=28997&parentSeq=29050
In partnership with
National Information Society Agency (NIA)
In partnership with
National Information Society Agency (NIA)
Connect with 500,000+ public servants solving your hardest challenges.
Connect with 500,000+ public servants solving your hardest challenges.
Help public servants worldwide learn from your work, what worked, what flopped and what you'd do differently
Share your project
Log in or sign up to continue the conversation