Smart Pothole Detection with Integrated Complaint Management System for Improving Road Safety and Maintenance

J. J. Seciliea, D. Shivani, B. Hemashree, Mykhailo Paslavskyi, Ahmed Samir Naje, Ijaz Ahad Department of Computer Science in AIML, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Department of Computer Science, Ukrainian National Forestry University, Lviv, Lviv Oblast, Ukraine. College of Engineering, Al-Qasim Green University, Al-Qasim, Babylon Governorate, Iraq. Department of Computer Systems Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan, China.

Abstract: Road damage, especially potholes, has been a major problem for many years and continues to affect our day-to-day lives. Such damage affects transportation and public safety and reduces road quality. Timely inspection and repair of such damage can help reduce these problems significantly and prevent accidents. However, manual road inspection consumes a huge amount of time. These manual methods can be inefficient, especially when covering a larger area is required. To address these issues, researchers have proposed a project to implement a smart pothole detection system with a complaint management system. This system uses deep learning to detect potholes and manage complaints. This system uses the YOLOv8 object detection model to identify potholes from road images. The dataset used to develop the model was collected from the Roboflow and Kaggle public dataset libraries for proper training and evaluation. This 1976-image dataset is divided into training, test, and validation to ensure proper learning and performance. A web-based interface was created using Streamlit. This web interface lets users upload pothole images. The trained model analyses these OpenCV-processed images. Boxes bind potholes in uploaded images. Also counted are potholes. The system lets people report road damage. Pandas stores all complaint data for analysis. This method maintains roads more effectively and faster than manual inspection, thereby improving public safety.

---

Keywords: Pothole Detection; Deep Learning (DL); Management System; Object Detection; Public Safety; Road Damage Detection; Complaint Management; Bounding Boxes.

---

Received on: 05/07/2025Revised on: 18/09/2025Accepted on: 27/10/2025Published on: 03/03/2026

---

---