The objective of this study is to propose and evaluate a method for automatically constructing a training dataset specific to each closed-circuit television (CCTV) camera to identify ice-road surface images using anomaly detection. In this approach, dry-road surface images were defined as normal, while ice-road surface images were considered as anomalous. By training the anomaly detection model solely on normal images, the system can identify ice-road surface images. However, the accuracy of identification may decline owing to variations in shooting locations, individual CCTV camera installation sites, and environmental conditions. To address these challenges, we used data from the Automated Meteorological Data Acquisition System (AMeDAS) near CCTV cameras to automatically select normal images, thereby enabling the construction of camera-specific training datasets. Specifically, if precipitation and temperature data from the nearest AMeDAS station—often located several kilometers away—consistently remained below predefined thresholds for a certain duration, the corresponding CCTV images were assumed to depict dry-road surfaces. We applied the proposed method to CCTV camera images and experimentally evaluated it using the PatchCore anomaly detection technique. The results demonstrated high accuracy, achieving an area under the receiver operating characteristic score of 0.961 and 0.999 during the daytime and nighttime, respectively, validating its effectiveness in detecting ice-road surface images. To demonstrate the practical viability of the proposed method, we validated it using footage from two additional CCTV cameras. The results showed that ice-road surface images could be easily identified using the proposed approach.

1. [1] A. Black and T. Mote, “Effects of Winter Precipitation on Automobile Collisions, Injuries, and Fatalities in the United States,” J. of Transport Geography, Vol.48, pp. 165-175, 2015. https://doi.org/10.1016/j.jtrangeo.2015.09.007
2. [2] Y. Zeng, Y. Qiang, N. Zhang, X. Yang, Z. Zhao, and X. Wang, “An Influencing Factors Analysis of Road Traffic Accidents Based on the Analytic Hierarchy Process and the Minimum Discrimination Information Principle,” Sustainability, Vol.16, No.16, Article No.6767, 2024. https://doi.org/10.3390/su16166767
3. [3] Ministry of Land, Infrastructure and Transport Website (in Japanese). https://www.mlit.go.jp/chosahokoku/h18giken/program/kadai/pdf/shitei/shi3-01.pdf [Accessed January 13, 2025]
4. [4] Ministry of Land, Infrastructure and Transport Website (in Japanese). https://www.mlit.go.jp/tec/it/denki/kikisiyou/touitusiyou_23dourokisyoukansokuR0403.pdf [Accessed September 18, 2024]
5. [5] Ministry of Land, Infrastructure and Transport Website (in Japanese). https://www.jma.go.jp/jma/kishou/books/jma-guidebook/chapter3.pdf [Accessed September 18, 2024]
6. [6] L. Colace, F. Santoni, and G. Assanto, “A Near-Infrared Optoelectronic Approach to Detection of Road Conditions,” Optics and Lasers in Engineering, Vol.51, No.5, pp. 633-636, 2013. https://doi.org/10.1016/j.optlaseng.2013.01.003
7. [7] J. Yu, J. Jiang, S. Fichera, P. Paoletti, L. Layzell, D. Mehta, and S. Luo, “Road Surface Defect Detection – From Image-Based to Non-Image-Based: A Survey,” arXiv:2402.04297, 2024. https://doi.org/10.48550/arXiv.2402.04297
8. [8] V. Viikari, T. Varpula, and M. Kantanen, “Road-Condition Recognition Using 24-GHz Automotive Radar,” IEEE Trans. on Intelligent Transportation Systems, Vol.10, No.4, pp. 639-648, 2009. https://doi.org/10.1109/TITS.2009.2026307
9. [9] M. Jokela, M. Kutila, and L. Le, “Road Condition Monitoring System Based on a Stereo Camera,” 2009 IEEE 5th Int. Conf. on Intelligent Computer Communication and Processing, pp. 423-428, 2009. https://doi.org/10.1109/ICCP.2009.5284724
10. [10] J. Moon and W. Park, “Using Support Vector Machines to Classify Road Surface Conditions to Promote Safe Driving,” Sensors, Vol.24, No.13, Article No.4307, 2024. https://doi.org/10.3390/s24134307
11. [11] Y. Nakamura, T. Hagiwara, S. Takahashi, Y. Nagata, R. Tsutamaki, and N, Matsuoka, “Investigation of Road Visibility Conditions using Precipitation Intensity measured by Weather Radar during Winter Seasons,” J. of the Eastern Asia Society for Transportation Studies, Vol.15, pp. 2395-2412, 2024. https://doi.org/10.11175/easts.15.2395
12. [12] Ministry of Land, Infrastructure and Transport Website (in Japanese). https://www.jma.go.jp/jma/kishou/know/amedas/kaisetsu.html [Accessed September 18, 2024]
13. [13] K. Ozcan, A. Sharma, S. Knickerbocker, J. Merickel, N. Hawkins, and M. Rizzo, “Road Weather Condition Estimation Using Fixed and Mobile Based Cameras,” Advances in Intelligent Systems and Computing, pp. 192-204, 2019. https://doi.org/10.1007/978-3-030-17795-9_14
14. [14] G. Pan, L. Fu, R. Yu, and M. Muresan, “Winter Road Surface Condition Recognition Using a Pre-trained Deep Convolutional Neural Network,” arXiv:1812.06858, 2018. https://doi.org/10.48550/arXiv.1812.06858
15. [15] A. Busch, D. Fink, M. Laves, Z. Ziaukas, M. Wielitzka, and T. Ortmaier, “Classification of Road Surface and Weather-Related Condition Using Deep Convolutional Neural Networks,” Advances in Dynamics of Vehicles on Roads and Tracks, pp. 1042-1051, 2019. https://doi.org/10.1007/978-3-030-38077-9_121
16. [16] K. Saito, M. Hutida, and A. Nakamura, “Basic Study of Frozen Road Surface Detection Method,” Dynamic Image Processing for Real Application Workshop, 2023 (in Japanese).
17. [17] S. Kawai, K. Takeuchi, K. Shibata, and Y. Horita, “A Smart Method to Distinguish Road Surface Conditions at Night-time using a Car-Mounted Camera,” IEEJ Trans. on Electronics, Information and Systems, Vol.134, No.6, pp. 878-884, 2014. https://doi.org/10.1541/ieejeiss.134.878
18. [18] I. Yamamoto, M. Kawana, I. Yamazaki, H. Tamura, and Y. Ookubo, “The Application of Visible Image Road Surface Sensors to Winter Road Management,” World Congress on intelligent Transport Systems, 2005.
19. [19] M. Rudolph, B. Wandt, and B. Rosenhahn, “Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flows,” IEEE/CVF Winter Conf. on Applications of Computer Vision, pp. 1906-1915, 2021. https://doi.org/10.1109/WACV48630.2021.00195
20. [20] Ministry of Land, Infrastructure and Transport Website (in Japanese). https://www.mlit.go.jp/tec/it/denki/kikisiyou/touitusiyou_01cctvR0303.pdf [Accessed September 18, 2024]
21. [21] A. Wang, H. Chen, L. Liu, K. Chen, Z. Lin, J. Han, and G. Ding, “YOLOv10: Real-Time End-to-End Object Detection,” arXiv:2405.14458, 2024. https://doi.org/10.48550/arXiv.2405.14458
22. [22] L. Bergman, N. Cohen, and Y. Hoshen, “Deep Nearest Neighbor Anomaly Detection,” arXiv:2002.10445, 2020. https://doi.org/10.48550/arXiv.2002.10445
23. [23] K. Roth, L. Pemula, J. Zepeda, B. Schölkopf, T. Brox, and P. Gehler, “Towards Total Recall in Industrial Anomaly Detection,” arXiv:2106.08265, 2022. https://doi.org/10.48550/arXiv.2106.08265