TY  - JOUR
T1  - A Computer Vision-Based Model for Road Marking and Road Damage Detection to Enhance Road Safety
AU  - Jang, Juhyeon 
AU  - Park, Myeongseok 
AU  - Kim, Jaeyun 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2024
DA  - 2024/1/1
DO  - 10.7840/kics.2024.49.5.653
KW  - Roadmark
KW  - road damage
KW  - road safety
KW  - U-net
KW  - YOLOv8
AB  - This paper addresses the challenges in road safety and infrastructure maintenance management posed by physical damages to road signs and road surfaces. Existing road monitoring methods that support these aspects often rely heavily on manual labor, proving to be inefficient, subjective, and yielding inconsistent results. In this paper, we propose a new model that dynamically detects road signs and damaged areas from the perspective of moving vehicles. This model integrates object detection technology for road sign recognition and segmentation technology for identifying damaged areas, forming a parallel relationship between these two stages. This approach not only assists in driving by identifying road damages but also extends to supporting maintenance, providing a fundamental solution compared to traditional systems. The proposed model focuses on real-time operation and maintenance support for damaged areas, offering a comprehensive solution for road safety and maintenance. By combining segmentation and object detection technologies, the proposed model aims to overcome the limitations of existing methods that either focus solely on real-time response or overlook various information about the road, aiming to provide a comprehensive solution for road safety and maintenance.