| Title |
Development of a Monitoring Device for a Camera-based Outdoor Marker Recognition Control System |
| Authors |
이석호(Seok-ho Lee) ; 장영민(Yeongmin Jang) ; 김지홍(Jeehong Kim) ; 정길도(Kil To Chong) |
| DOI |
https://doi.org/10.5573/ieie.2025.62.3.93 |
| Keywords |
Automated lane painting; Embedded; Marker recognition; Lane detection; Center of mass |
| Abstract |
Detecting the position of outdoor markers as like lanes from camera image processing and using it for precise control is critical in automated control systems. However, existing outdoor marker recognition techniques have poor recognition accuracy in worn out aging road environments, deep learning techniques require large amounts of training data, and trained models have limitations in accurately calculating position errors in new environments. In this paper, we developed an observation system that provides real-time feedback on the position error between markers. Based on an embedded board and a high-resolution camera, the system measures the position error between markers in real time and generates a feedback signal to the control unit. The experimental results showed a recognition rate of 94.046% in the pre-recorded video test, and a high recognition rate of 98.165% in the field test. This camera-based feedback signal generation system can be utilized for position control by giving appropriate compensation values to the control device. |