YOLOv8 기반 딥러닝 모델을 이용한 동적 교통객체 검출 성능 비교 분석 연구
A Comparative Study on Dynamic Traffic Object Detection Performance Using YOLOv8-Based Deep Learning Model
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 연구는 교통사고 예방 목적의 CCTV 관제센터 기반 동적객체 감지 기술 실현 및 실용화를 위한 기초연구로 YOLOv8을 활용하여 다양한 실험조건에서의 교통객체 감지율을 비교분석하였다. 전체 학습 데이터셋의 mAP50은 0.952로 실용화가 기대되는 안정적인 감지율을 나타냈다. 8차선 학원가사거리의 RGB CCTV 영상에서 주간의 전체 감지율이 야간보다 높게 분석되었으며, 도로규모별 분석결과는 8차선 도로가 2차선 도로보다 객체 감지율이 떨어졌고, 8차선도로에서 보행자 감지는 차량에 비해 예측력이 낮게 분석되었다. 학원가사거리 횡단보도에서군집보행과 다량·다종의 모빌리티가 동시다발적으로 나타나는 특징 때문으로 판단된다. 영상유형별 분석결과, 열화상보다 높은 예측력이 확보된 RGB 영상이 전반적인 교통객체 관제에서적합한 것으로 나타났다. 향후 기술 실현 및 실용화를 위한 후속연구가 지속적으로 진행되어국민안전을 위한 지능형 교통관제가 이루어지길 기대한다.
Abstract
This study compares traffic object detection rates under various experimental conditions using YOLOv8. The mAP50 of the entire learning dataset was 0.952, indicating a stable detection rate that is expected to be put into practical use. In RGB CCTV images of an eight-lane academy street, the overall detection rate during the day was found to be higher than at night. Based on each road size, an eight-lane main road has a lower object detection rate than a two-lane side road, and pedestrians on the eight-lane road had lower predictive power than vehicles. This is believed to be due to the characteristics of walking in clusters at a crosswalk on main roads, simultaneously showing various and large amounts of mobility. From each image type, RGB showed higher predictive power than thermal images, and accordingly, RGB images are considered advantageous for overall traffic object control. It is hoped that follow-up research for the realization and practical use of technology will continue in the future, and through this, intelligent traffic control for public safety will be achieved.
참고문헌 (25)
- Bak, G. T., "Performance evaluation of deep learning-based nighttime object detection using histogram equalization", Journal of the Korea Academia-Industrial cooperation Society, 25(2), pp.52~61 (2024)
- Chang, J. U., "Thermal imagery-based object detection algorithm for low-light level nighttime surveillance system", The Journal of The Korea Institute of Intelligent Transportation Systems, 19(3), pp.129~136 (2020)
- Choi, W., "A study on the establishment of urban life safety abnormalities detection service using multi-type complex sensor information", Journal of the Society of Disaster Information, 20(2), pp.315~328 (2024)
- Choi, W., "A study on selection of bicycle road hazard detection elements for mobile IoT sensor device operation", The Journal of The Korea Institute of Intelligent Transportation Systems, 23(4), pp.37~53 (2024)
- Choi, W., "Evaluation of preferences for a thermal-camera-based abnormal situation detection service via the integrated fuzzy AHP", TOPSIS model, 2023, pp.13, 11591~
- González, A., "Pedestrian detection at day", night time with visible and FIR cameras: A comparison, 2016, pp.16(6), 820~
- Hwang, S. Y., "Development for analysis service of crowd density in CCTV video using YOLOv4", The Journal of The Institute of Internet, Broadcasting and Communication, 24(3), pp.177~182 (2024)
- Hwang, W., "A study on the development of night vision thermal camera using deep learning", The Journal of Korean Institute of Communications and Information Sciences, 44(12), pp.2332~2338 (2019)
- Kim, J. S., "Analysis of building object detection based on the YOLO neural network using UAV images", Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 39(6), pp.381~392 (2021)
- Ko, A., "Comparative analysis of YOLOv8 object detection model performance in fire detection in traditional markets using thermal cameras", Journal of Korea Society of Digital Industry and Information Management, 19(4), pp.117~126 (2023)
- "KOSIS(Korean Statistical Information Service)" (2024)[URL]
- Kwon, K., "A study on occupancy prediction using low-resolution thermal images for efficient building energy management", Journal of Energy Engineering, 33(2), pp.43~57 (2024)
- Lee, C. E., "Development of fire detection system using YOLOv8", Journal of the Semiconductor & Display Technology, 23(1), pp.19~24 (2024)
- Lee, H. D., "A study on traffic vulnerable detection using object detection-based ensemble and YOLOv5", Journal of The Korea Society of Computer and Information, 29(1), pp.61~68 (2024)
- Lee, J. M., "Improvement of negative emotion recognition in visible images enhanced by thermal imaging", Sustainability, 14(22), pp.15200~ (2022)
- Lee, N. Y., "Designing a smart safe transportation system within a university using object detection algorithm", Journal of The Korea Society of Computer and Information, 29(1), pp.51~59 (2024)
- Lee, T. H., "Real-time fire detection method using YOLOv8", Journal of the Semiconductor & Display Technology, 22(2), pp.77~80 (2023)
- Lee, Y. H., "Comparison of CNN and YOLO for object detection", Journal of the Semiconductor & Display Technology, 19(1), pp.85~92 (2020)
- Miah, M. S., "E-scooter detection in media using deep learning", Journal of Contents Computing, 4(1), pp.393~400 (2022)
- Munir, F., "Exploring thermal images for object detection in underexposure regions for autonomous driving", Applied Soft Computing, 121, pp.108793~ (2022)
- NamGung, D. H., "Proposal of new labeling method for detection improvement of two-people riding on E-scooter in YOLOv8n", Journal of Digital Contents Society, 24(7), pp.1575~1581 (2023)
- Pavez, V., "Thermal image generation for robust face recognition", Applied Sciences, 12(1), pp.497~ (2022)
- Seo, J., "Application of deep learning-based object detection and distance estimation algorithms for driving to urban area", The Journal of The Korea Institute of Intelligent Transportation Systems, 21(3), pp.83~95 (2022)
- Wang, B., "Behavioral parameter field for human abnormal behavior recognition in low-resolution thermal imaging video", Applied Sciences, 12(1), pp.402~ (2022)
- Yun, B., "A study on improvement of dynamic object detection using dense grid model and anchor model", The Journal of The Korea Institute of Intelligent Transportation Systems, 17(3), pp.98~110 (2018)