A Low-Cost Smartphone-Based Photogrammetric Approach for Highway Pavement Defects Detection and Mapping

Abstract

Evaluating the condition of the pavement is a crucial step before undertaking any maintenance or rehabilitation work. This study presents an innovative low-cost smartphone-based workflow that leverages photogrammetric techniques for mapping and measuring pavement surface defects, specifically designed to overcome the limitations of existing UAV (Unmanned Aerial Vehicle) and LiDAR (Light Detection and Ranging) based inspection methods, including high costs, operational complexity, and regulatory or environmental constraints. Unlike existing smartphone-based approaches that primarily focus on defect detection or classification, this method enables both accurate geolocation and precise measurement of key defect dimensions (length, width, and area), generating georeferenced frames interpretable through Geographic Information Systems (GIS). The workflow relies on a Redmi 9 smartphone and its integrated sensors, including camera, accelerometer, gyroscope, magnetometer and GPS (Global Positioning System) combining video capture and GPS/IMU (IMU, Inertial Measurement Unit) recording to produce georeferenced data suitable for GIS analysis. The system was tested on a section of the East-West Highway (from Kilometer Point 150 to 193, in Mila and Constantine provinces, Algeria), driven at a cruising speed of 70 km/h. Compared to a reference survey conducted in situ with a Global Navigation Satellite System (GNSS) receiver (Garmin GPSmap 62s), the results demonstrated high performance, with an F1 score of 93.27%. The system proved effective in detecting and mapping the main surface defects, such as longitudinal and transverse cracks, alligator cracking, and rutting. This novel smartphone-based workflow provides a cost-effective, rapid, and operationally flexible solution for road maintenance planning, enabling targeted and optimized interventions. Particularly useful in developing countries, this method allows for cyclic inventories of road infrastructure with a modest budget while maintaining reliability and accessibility.