Ver PDF
Registro

Inventario de baches y mantenimiento predictivo de pavimentos mediante visión computacional

Montúfar, Gabriel

Pothole inventory and predictive pavement maintenance using computer vision



Autores Montúfar, Gabriel
Tema Artificial intelligence
Tema Computer vision
Tema Detection
Tema Maintenance
Tema Roads
Tema Carreteras
Tema Detección
Tema Inteligencia artificial
Tema Mantenimiento
Tema Visión por ordenador
Descripción Road infrastructure is essential for economic and social development, but potholes pose a significant challenge affecting safety, comfort, and maintenance costs. This article explores the use of computer vision and deep learning for automated pothole inventory and predictive pavement maintenance. Methods based on convolutional neural networks, such as dilated convolutions and object detection models, are reviewed to identify defects in real time with high accuracy. Data acquisition techniques using aerial, satellite, and mobile device images are analyzed, integrating algorithms for segmentation and damage classification. The predictive approach uses historical data and multitask models to estimate pavement condition indices, enabling timely interventions and cost reduction. Limitations such as low-light conditions and image resolution are discussed, proposing improvements through data augmentation and sensor fusion. Results from recent studies show accuracies above 90 percent, highlighting the potential for intelligent transportation systems. This work contributes to the sustainability of road networks in urban and rural contexts, promoting efficient management based on emerging technologies.
Descripción La infraestructura vial es esencial para el desarrollo económico y social, pero los baches representan un desafío significativo que afecta la seguridad, el confort y los costos de mantenimiento. Este artículo explora el uso de la visión computacional y el aprendizaje profundo para el inventario automatizado de baches y el mantenimiento predictivo de pavimentos. Se revisan métodos basados en redes neuronales convolucionales, como convoluciones dilatadas y modelos de detección de objetos, que permiten identificar defectos en tiempo real con alta precisión. Se analizan técnicas de adquisición de datos mediante imágenes aéreas, satelitales y dispositivos móviles, integrando algoritmos para segmentación y clasificación de daños. El enfoque predictivo utiliza datos históricos y modelos multitarea para estimar el índice de condición de pavimentos, facilitando intervenciones oportunas y reduciendo costos. Se discuten limitaciones como condiciones de iluminación baja y resolución de imágenes, proponiendo mejoras mediante aumento de datos y fusión de sensores. Los resultados de estudios recientes muestran precisiones superiores al 90 por ciento, destacando el potencial para sistemas inteligentes de transporte. Este trabajo contribuye a la sostenibilidad de las redes viales en contextos urbanos y rurales, promoviendo una gestión eficiente basada en tecnologías emergentes.
Editorial Universidad Metropolitana de Educación, Ciencia y Tecnología
Fecha 2025-12-31
Tipo info:eu-repo/semantics/article
Tipo info:eu-repo/semantics/publishedVersion
Formato application/pdf
Formato application/epub+zip
Identificador https://revistas.umecit.edu.pa/index.php/oratores/article/view/1844
Identificador 10.37594/oratores.n23.1844
Fuente Oratores Journal; No. 23 (2025): ORATORES December 2025- May 2026; 167-181
Fuente Revista Oratores; Núm. 23 (2025): ORATORES Diciembre 2025 - Mayo 2026; 167-181
Fuente 2644-3988
Fuente 2410-8928
Fuente 10.37594/oratores.n23
Idioma spa
Relación https://revistas.umecit.edu.pa/index.php/oratores/article/view/1844/2804
Relación https://revistas.umecit.edu.pa/index.php/oratores/article/view/1844/2805
Relación Ahmed, K. R. (2021). Smart pothole detection using deep learning based on dilated convolution. Sensors, 21(24), 8406. https://doi.org/10.3390/s21248406
Relación Asad, M. H., Khaliq, S., Yousaf, M. H., Ullah, M. O., & Ahmad, A. (2022). Pothole detection using deep learning: A real-time and AI-on-the-edge perspective. Advances in Civil Engineering, 2022, Article 9221211. https://doi.org/10.1155/2022/9221211
Relación Dipankar, A., & Suman, S. K. (2024). Pavement Distress Detection Based on the Staged Deep Learning Classifier Approach. Transportation Research Record, 2679(2), 1015-1033. https://doi.org/10.1177/03611981241265689
Relación Han, S., Jiang, Y., Uwakweh, B., & Zhang, K. (2023). PCIer: Pavement condition evaluation using aerial imagery and deep learning. Geographies, 3(1), 132–142. https://doi.org/10.3390/geographies3010008
Relación Ibragimov, E., Kim, N., Lee, S., Cho, N., & Lee, J. (2024). Automated pavement condition index assessment with deep learning. Sensors, 24(7), 2333. https://doi.org/10.3390/s24072333
Relación Khandakar, A., Michelson, D.G., Naznine, M. et al. Harnessing Smartphone Sensors for Enhanced Road Safety: A Comprehensive Dataset and Review. Sci Data 12, 418 (2025). https://doi.org/10.1038/s41597-024-04193-0
Relación Kim, G., & Kim, S. (2024). A Road Defect Detection System Using Smartphones. Sensors, 24(7), 2099. https://doi.org/10.3390/s24072099
Relación Kim, S., Seo, D., & Jeon, S. (2023). Improvement of tiny object segmentation accuracy in aerial images for asphalt pavement pothole detection. Sensors, 23(13), 5851. https://doi.org/10.3390/s23135851
Relación Kim, Y.-M., Kim, Y.-G., Son, S.-Y., Lim, S.-Y., Choi, B.-Y., & Choi, D.-H. (2022). Review of recent automated pothole-detection methods. Applied Sciences, 12(11), 5320. https://doi.org/10.3390/app12115320
Relación Lebaku, P. K. R., Gao, L., Lu, P., & Sun, J. (2024). Deep learning for pavement condition evaluation using satellite imagery. Infrastructures, 9(9), 155. https://doi.org/10.3390/infrastructures9090155
Relación Li, D., Duan, Z., Hu, X., Zhang, D., & Zhang, Y. (2023). Automated classification and detection of multiple pavement distress images based on deep learning. Journal of Traffic and Transportation Engineering (English Edition), 10(3), 436-451. https://doi.org/10.1016/j.jtte.2021.04.008
Relación Ma, N., Fan, J., Wang, W., Wu, J., Jiang, Y., Xie, L., & Fan, R. (2022). Computer vision for road imaging and pothole detection: A state-of-the-art review of systems and algorithms. Transportation Safety and Environment, 4(4), tdac026. https://doi.org/10.1093/tse/tdac026
Relación Owor, N. J., Du, H., Daud, A., Aboah, A., & Adu-Gyamfi, Y. (2023). Image2PCI—A multitask learning framework for estimating pavement condition indices directly from images. arXiv. https://arxiv.org/abs/2310.08538
Relación Park, S.-S., & Nguyen, N.-N. (2025). Two-camera vision technique for measuring pothole area and depth. Measurement, 247(Suppl C), Article 116809. https://doi.org/10.1016/j.measurement.2025.116809
Relación Ruseruka, C., Mwakalonge, J., Comert, G., Siuhi, S., Ngeni, F. C., & Anderson, Q. (2023, December 6). Augmenting roadway safety with machine learning and deep learning: Pothole detection and dimension estimation using in-vehicle technologies [Preprint]. SSRN. https://doi.org/10.2139/ssrn.4655731
Relación Ruseruka, C., Mwakalonge, J., Comert, G., Siuhi, S., Ngeni, F., & Anderson, Q. (2024). Augmenting roadway safety with ML and DL: Pothole detection and dimension estimation using in-vehicle technologies. Machine Learning with Applications, 16, 100547. https://doi.org/10.1016/j.mlwa.2024.100547
Relación Samadzadegan, F., Arefi, H., & Safari, A. (2024). Automatic pavement distress recognition using UAV imagery and YOLOv8. Drones, 8(6), 244. https://www.mdpi.com/2504-446X/8/6/244
Relación Setayesh Valipour, P., Golroo, A., Kheirati, A., & Moghaddasi, M. (2024). Automatic pavement distress severity detection using deep learning. Road Materials and Pavement Design, 25(8), 1830–1846. https://doi.org/10.1080/14680629.2023.2276422
Relación Yang, H., Cao, J., Wan, J. et al. A Large-Scale Image Repository for Automated Pavement Distress Analysis and Degradation Trend Prediction. Sci Data 12, 1426 (2025). https://doi.org/10.1038/s41597-025-05748-5
Relación Yang, X., Zhang, J., Liu, W., Jing, J., Zheng, H., & Xu, W. (2024). Automation in road distress detection, diagnosis and treatment. Journal of Road Engineering. https://doi.org/10.1016/j.jreng.2024.01.005
Relación Zanevych, Y., Yovbak, V., Basystiuk, O., Shakhovska, N., Fedushko, S., & Argyroudis, S. (2024). Evaluation of pothole detection performance under low-light conditions. Sustainability, 16(24), 10964. https://doi.org/10.3390/su162410964
Relación Zhang, H., Dong, Y., Hou, Y., Cheng, X., Xie, P., & Di, K. (2025). Research on Asphalt Pavement Surface Distress Detection Technology Coupling Deep Learning and Object Detection Algorithms. Infrastructures, 10(4), 72. https://doi.org/10.3390/infrastructures10040072
Relación Zhang, Z., Song, W., Zhuang, Y., Zhang, B., & Wu, J. (2024). Automated Multi-Type Pavement Distress Segmentation and Quantification Using Transformer Networks for Pavement Condition Index Prediction. Applied Sciences, 14(11), 4709. https://doi.org/10.3390/app14114709
Relación Zheng, L., Xiao, J., Wang, Y., et al. (2024). Deep learning-based intelligent detection of pavement distress. Automation in Construction, 168, 105772. https://doi.org/10.1016/j.autcon.2024.105772
Relación Zuo, H., Li, Z., Gong, J., & Tian, Z. (2025, April 16). Intelligent road crack detection and analysis based on improved YOLOv8 [Preprint]. arXiv. https://doi.org/10.48550/arXiv.2504.13208
Derechos Derechos de autor 2025 Revista ORATORES
Derechos https://creativecommons.org/licenses/by-nc-sa/4.0