All Issue

2025 Vol.58, Issue 11 Preview Page

Research Article

A study on real-time object detection-based disaster type recognition and classification

실시간 객체 인식 기반 재난 유형 탐지 및 분류에 관한 연구

Yejin Leea
,
Su Han Namb
,
Sung Eun Kimc
,
Young Do Kimd*
이 예진a
,
남 수한b
,
김 성은c
,
김 영도d*
aMaster’s Course, Department of Civil & Engineering, Myongji University, Yongin, Korea
bPh.D. Course, Department of Civil & Engineering, Myongji University, Yongin, Korea
cPh.D, Urban Environment & Climate Change Research Division, The Seoul Institute, Seoul, Korea
dProfessor, Division of smart Infrastructure Engineering, University, Yongin, Korea
a명지대학교 토목환경공학과 석사과정
b명지대학교 토목환경공학과 박사과정
c서울연구원 지속가능연구실 연구위원
d명지대학교 스마트인프라공학부 교수
*Corresponding Author
30 November 2025. pp. 1057-1068
PDF XML Citation
Abstract

Disasters occur in diverse and unpredictable forms, underscoring the necessity of rapid initial response. Citizen-reported images captured via smartphones provide a valuable means of acquiring disaster scene information without temporal or spatial constraints; however, the risk of misinformation necessitates careful verification. This study developed a YOLOv8-based object detection model to detect and classify six disaster types forest fire, vehicle fire, building fire, pothole, road crack, and flood using citizen-reported images. The dataset was collected, preprocessed, and augmented, and transfer learning was applied to optimize model performance. Evaluation was conducted through cross-validation and an independent test dataset, yielding strong results with Precision = 0.956, Recall = 0.937, mAP@0.5 = 0.964, and mAP@0.5-0.95 = 0.792. The model achieved excellent performance in fire- and flood-related categories, while relatively lower performance was observed for forest fire and road crack due to similarities in color and texture with the background. The proposed model improves the reliability and timeliness of image-based disaster detection and classification, contributing to the advancement of early disaster response and management systems. Furthermore, it supports the verification of citizen-reported images, reducing false detections caused by misinformation, and provides a foundation for building standardized disaster image databases and developing intelligent disaster management systems.

Keywords
YOLOv8
Object detection
Disaster classification
Citizen-reported images
AI-based disaster management

재난은 자연적 또는 인위적 요인으로 인해 예측이 어렵게 발생하며, 복합적 양상으로 확대되어 심각한 피해를 초래한다. 이러한 재난 상황에서 신속하고 정확한 탐지는 피해 최소화를 위한 초기 대응의 핵심 요소이다. 본 연구에서는 6종 재난 유형을 탐지·분류할 수 있는 YOLOv8 기반 객체 탐지 모델을 개발하였다. 수집된 이미지를 전처리하고 애노테이션을 수행한 후, 다양한 이미지 증강(image augmentation) 기법을 적용하여 모델의 일반화 성능을 강화하였다. 성능 평가 결과 Precision = 0.956, Recall = 0.937, mAP@0.5 = 0.964, mAP@0.5–0.95 = 0.792로 높은 탐지 정확도를 보였다. 특히 화재 및 홍수 관련 재난에서 우수한 성능을 나타냈으나, 산불화재 및 도로균열처럼 배경과 색상·질감이 유사한 객체에서는 상대적으로 낮은 탐지율을 보였다. 제안된 모델은 다양한 환경에서도 신속하고 정확한 재난 탐지 및 분류가 가능한 고성능 프레임워크를 제시하였으며, 향후 시민제보, 드론 영상, CCTV 영상 등 실시간 영상 데이터와 연계한 재난 대응 및 관리 체계의 고도화에 기여할 수 있을 것으로 기대된다.

키워드
YOLOv8
객체 인식
재난분류
시민제보 이미지
인공지능 기반 재난 관리
References
1

Ahn, H., Kim, J.I., and Kim, T. (2014). “3D accuracy analysis of mobile phone-based stereo images.” Journal of Broadcast Engineering, Vol. 19, No. 5, pp. 677-686.

10.5909/JBE.2014.19.5.677
2

Antzoulatos, G., Giannakeris, P., Koulalis, I., Karakostas, A., Vrochidis, S., and Kompatsiaris, I. (2020). “A multi-layer fusion approach for real-time fire severity assessment based on multimedia incidents.” In Proceedings of the 17th International Conference on Information Systems for Crisis Response and Management - ISCRAM 2020, Blacksburg, VA, U.S., pp. 24-27.

3

Asif, A., Khatoon, S., Hasan, M.M., Alshamari, M.A., Abdou, S., Elsayed, K.M., and Rashwan, M. (2021). “Automatic analysis of social media images to identify disaster type and infer appropriate emergency response.” Journal of Big Data, Vol. 8, 83.

10.1186/s40537-021-00471-5
4

Bhattacharjee, S., Roy, S., and Bit, S.D. (2019). “Post-disaster map builder: Crowdsensed digital pedestrian map construction of the disaster affected areas through smartphone based DTN” Computer Communications, Vol. 134, No. 15, pp. 96-113.

10.1016/j.comcom.2018.11.010
5

Cheng, G., Chen, X., Wang, C., Li, X., Xian, B., and Yu, H. (2024). “Visual fire detection using deep learning: A survey.” Neurocomputing, Vol. 596, 127975.

10.1016/j.neucom.2024.127975
6

Choi, M., Starbuck, R., Lee, S., Hwang, S., Lee, S., Park, M., and Lee, H.S. (2018). “Distributed and interoperable simulation for comprehensive disaster response management in facilities.” Automation in Construction, Vol. 93, pp. 12-21.

10.1016/j.autcon.2018.05.007
7

Cicek, D., and Kantarci, B. (2023). “Use of mobile crowdsensing in disaster management: A systematic review, challenges, and open issues.” Sensors, Vol. 23, No. 3, 1699. doi: 10.3390/s23031699.

10.3390/s2303169936772738PMC9918985
8

Daly, S., and Thom, J.A. (2016). “Mining and classifying image posts on social media to analyse fires.” 13th International Conference on Information Systems for Crisis Response and Management (ISCRAM 2016), Rio de Janeiro, Brazil, pp. 1-14.

9

Geiß, M., Wagner, R., Baresch, M., Steiner, J., and Zwick, M. (2023). “Automatic bounding box annotation with small training datasets for industrial manufacturing.” Micromachines, Vol. 14, No. 2, 442.

10.3390/mi1402044236838142PMC9962188
10

Gragnaniello, D., Greco, A., Sansone, C., and Vento, B. (2024). “Fire and smoke detection from videos: A literature review under a novel taxonomy.” Expert Systems with Applications, Vol. 255, 124783.

10.1016/j.eswa.2024.124783
11

Guerrini, C.J., Majumder, M.A., Lewellyn, M.J., and McGuire, A.L. (2018). “Citizen science, public policy.” Science, Vol. 361, No. 6398, pp. 134-136.

10.1126/science.aar837930002244PMC6534494
12

Irwin, A. (2018). “No PhDs needed: How citizen science is transforming research.” Nature, Vol. 562, No. 7726, pp. 480-483.

10.1038/d41586-018-07106-5
13

Kim, H.S., Choi, K.H., and Kim, C.J. (2023). “Disaster site location determination using smartphones and accuracy analysis.” Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, Vol. 41, No. 6, pp. 559-565.

10.7848/ksgpc.2023.41.6.559
14

Kim, M., and Kang, J. (2022). “Exploring spatial distributions of temporary flood shelters in Seoul.” The Geographical Journal of Korea, Vol. 56, No. 3, pp. 235-244.

10.22905/kaopqj.2022.56.3.7
15

Kim, S., and Ahn, J. (2024), “A YOLO-based crop pests detection mobile application for smart farming.” Journal of the Korea Academia-Industrial Cooperation Society, Vol. 25, No. 7, pp. 603-610.

10.5762/KAIS.2024.25.7.603
16

Lee, H.J., Yun, H.S., and Han, H. (2021), “A study on the selection of types of social disasters by region” Journal of the Society of Disaster Information, Vol. 17, No. 2, pp. 206-217.

17

Lee, J.E., and Lee, W.K. (2014), “Hybrid disaster response and crisis management system in Korea.” Crisisonomy, Vol. 10, No. 9, pp. 15-31.

18

Lee, S. (2019). “A framework for climate change risk management.” Journal of the Society of Disaster Information, Vol. 15, No. 3, pp. 367-379.

19

Lee, S.H., and Kim, M.S. (2019). “Establishment of a disaster response support system based on public-private partnerships - Focused on cooperation with a private institution using drone.” Crisisonomy, Vol. 15, No. 11, pp. 1-15.

10.14251/crisisonomy.2019.15.11.1
20

Lukyanenko, R., Parsons, J., and Wiersma, Y.F. (2016). “Emerging problems of data quality in citizen science.” Conservation Biology, Vol. 30, No. 3, pp. 447-449.

10.1111/cobi.12706
21

Na, Y., Park, J., and Choi, J. (2019). “Analysis of the characteristics of a complex disaster using bigdata.” The Geographical Journal of Korea, Vol. 53, No. 4, pp. 375-384.

22

Oksuz, K., Cam, B.C., Akbas, E., and Kalkan, S. (2018). “Localization recall precision (LRP): A new performance metric for object detection.” In Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, pp. 504-519.

10.1007/978-3-030-01234-2_31
23

Paul, J.D., Buytaert, W., Allen, S., Ballesteros-Cánovas, J.A., Bhusal, J., Cieslik, K., Clark, J., Dugar, S., Hannah, D.M., Stoffel, M., Dewulf, A., Dhital, M.R., Liu, W., Nayaval, J.L., Neupane, B., Schiller, A., Smith, P.J., and Supper, R. (2018). “Citizen science for hydrological risk reduction and resilience building.” Wiley Interdisciplinary Reviews-Water, Vol. 5, No. 1, e1262.

10.1002/wat2.1262
24

Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016), “You only look once: Unified, real-time object detection.” In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, U.S., pp. 779-788.

10.1109/CVPR.2016.91
25

Saito, T., and Rehmsmeier, M. (2015). “The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets.” PloS One, Vol. 10, No. 3, e0118432.

10.1371/journal.pone.011843225738806PMC4349800
26

Saleh, A., Zulkifley, M.A., Harun, H.H., Gaudreault, F., Davison, I., and Spraggon, M. (2024). “Forest fire surveillance systems: A review of deep learning methods.” Heliyon, Vol. 10, No. 1, e23127.

10.1016/j.heliyon.2023.e2312738163175PMC10754902
27

Santos Junior, E.S.D., Paixao, T., and Alvarez, A.B. (2025). “Comparative performance of YOLOv8, YOLOv9, YOLOv10, and YOLOv11 for layout analysis of historical documents images.” Applied Sciences, Vol. 15, No. 6, 3164.

10.3390/app15063164
28

Sharma, A., Kumar, V., and Longchamps, L. (2024). “Comparative performance of YOLOv8, YOLOv9, YOLOv10, YOLOv11 and Faster R-CNN models for detection of multiple weed species.” Smart Agricultural Technology, Vol. 9, 100648.

10.1016/j.atech.2024.100648
29

Terven, J., Cordova-Esparza, D.-M., and Romero-Gonzalez, J.-A., (2023). “A comprehensive review of YOLO architectures in computer vision: From YOLOv1 to YOLOv8 and YOLO-NAS.” Machine Learning and Knowledge Extraction, Vol. 5, No. 4, pp. 1680-1716.

10.3390/make5040083
30

Wu, S., Sheng, B., Fu, G., Zhang, D., and Jian, Y. (2024). “Multiscale fire image detection method based on CNN and Transformer.” Multimedia Tools and Applications, Vol. 83, No. 16, pp. 49787-49811.

10.1007/s11042-023-17482-4
31

Yuan, Q., Shi, Y., and Li, M. (2024). “A review of computer vision-based crack detection methods in civil infrastructure: Progress and challenges.” Remote Sensing, Vol. 16, No. 16, 2910.

10.3390/rs16162910
Information
  • Publisher :KOREA WATER RESOURECES ASSOCIATION
  • Publisher(Ko) :한국수자원학회
  • Journal Title :Journal of Korea Water Resources Association
  • Journal Title(Ko) :한국수자원학회 논문집
  • Volume : 58
  • No :11
  • Pages :1057-1068
  • Received Date : 2025-08-18
  • Revised Date : 2025-10-12
  • Accepted Date : 2025-10-13
  • DOI :https://doi.org/10.3741/JKWRA.2025.58.11.1057
Journal Informaiton Journal of Korea Water Resources Association Journal of Korea Water Resources Association
Online Submission
Archives
  • scopus
  • KCI-수자원
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close