Improving Image Segmentation under Adverse Conditions in Coastal Infrastructure Monitoring

• Pedro V. Rubinstein UFRJ http://orcid.org/0009-0008-4376-383X
• Ricardo L. Kauer UFRJ http://orcid.org/0009-0002-4289-2345
• Alexandre G. Cardeman COR
• Marcelo Abelheira COR
• Pedro Cruz UFRJ https://orcid.org/0000-0002-2019-133X
• Rodrigo S. Couto UFRJ https://orcid.org/0000-0002-6921-7756
• Luís Henrique M. K. Costa UFRJ https://orcid.org/0000-0002-7789-3359

DOI: https://doi.org/10.5753/courb.2025.9518

Abstract

Smart cities increasingly rely on AI-driven solutions. In coastal areas like Rio de Janeiro, maritime risks present significant challenges, as exemplified by the collapse of the Tim Maia Bike Path. This paper proposes a method for monitoring coastal infrastructure using a custom segmentation model based on YOLO segmentation, reducing the need for constant human supervision. However, external camera placement introduces challenges making segmentation difficult. To address these challenges, we investigate the effects of data augmentation on AI model performance. As a case study, we apply this method to develop a system for the Tim Maia Bike Path, with models achieving 98% mAP50-95 throughout the day.

Keywords: Smart Cities, Urban Computing for Protection and Public Safety, Anomaly Detection and Event Detection in Urban Areas, Environmental Protection with Urban Computing

References

Alam, M. J., Hasan, M. Z., Rahman, M. M., Rahman, M. A., Sarker, N. H., Azad, S., Islam, T. N., Paul, B., Anjum, T., Halder, B., & Fattah, S. A. (2025). An optimized YOLOv5 based approach for real-time vehicle detection at road intersections using fisheye cameras.

CVAT.ai Corporation. (2023). Computer Vision Annotation Tool (CVAT).

Herath, H., & Mittal, M. (2022). Adoption of artificial intelligence in smart cities: A comprehensive review. International Journal of Information Management Data Insights, 2(1), 100076.

Jocher, G., Qiu, J., & Chaurasia, A. (2023). Ultralytics YOLO.

Kaur, R., Roul, R. K., & Batra, S. (2023). A hybrid deep learning CNN-ELM approach for parking space detection in smart cities. Neural Computing and Applications, 35(18), 13665–13683.

Liu, X. (2024). The application of CNN in smart city management and supervision under the background of smart city. Measurement: Sensors, 33, 101227.

Martinho, L., Neto, O., Calvalcanti, J., Pio, J., & Oliveira, F. (2023). An approach for fish detection in underwater images. In Anais do XVIII Workshop de Visão Computacional (pp. 6–11). Porto Alegre, RS, Brasil: SBC.

Midigudla, R. S., Dichpally, T., Vallabhaneni, U., Wutla, Y., Sundaram, D. M., & Jayachandran, S. (2025). A comparative analysis of deep learning models for waste segregation: YOLOv8, EfficientDet, and Detectron 2. Multimedia Tools and Applications.

Muhadi, N. A., Abdullah, A. F., Bejo, S. K., Mahadi, M. R., & Mijic, A. (2021). Deep learning semantic segmentation for water level estimation using surveillance camera. Applied Sciences, 11(20).

Onufriichuk, O. (2025). Aiogram.

Pachouly, J., Ahirrao, S., Kotecha, K., Selvachandran, G., & Abraham, A. (2022). A systematic literature review on software defect prediction using artificial intelligence: Datasets, data validation methods, approaches, and tools. Engineering Applications of Artificial Intelligence, 111, 104773.

Rio, G. (2016). Parte da ciclovia desaba em São Conrado, zona sul do Rio.

Scardino, G., Scicchitano, G., Chirivì, M., Costa, P. J. M., Luparelli, A., & Mastronuzzi, G. (2022). Convolutional neural network and optical flow for the assessment of wave and tide parameters from video analysis (Leucotea): An innovative tool for coastal monitoring. Remote Sensing, 14(13).

Talaat, F. M., & ZainEldin, H. (2023). An improved fire detection approach based on YOLOv8 for smart cities. Neural Computing and Applications, 35(28), 20939–20954.

Telegram. (2025). Telegram.

Wang, J., & Zhao, H. (2024). Improved YOLOv8 algorithm for water surface object detection. Sensors, 24(15).

Yuldashev, Y., Mukhiddinov, M., Abdusalomov, A. B., Nasimov, R., & Cho, J. (2023). Parking lot occupancy detection with improved MobileNetV3. Sensors, 23(17).

Zaib Ullah, F., Al-Turjman, L. M., & Gagliardi, R. (2020). Applications of artificial intelligence and machine learning in smart cities. Computer Communications, 154, 313-323.