Road Surface Quality Detection using Smartphone Sensors: Egyptian Roads Case Study

Abstract

Road anomalies have significant negative effects on both passengers and vehicles such as traffic congestion and accidents. Nowadays, smartphones are ubiquitous and used by so many drivers, at least to know the driving directions to their destination. Several studies utilized this observation and used the smartphone embedded sensors to detect road anomalies. In this paper, we evaluate the effectiveness of this methodology with sensor readings obtained while driving in Egyptian roads. An android application is developed to record sensor readings while driving over the road anomalies. Four datasets are collected for different streets in Cairo of total duration of 80 minutes and about 50K records. To automatically label these datasets, two clustering techniques (K-Means and DBSCAN) are evaluated to give the ground truth for the sensor readings if they represent road anomalies or normal road surface. It is noticed that DBSCAN can accurately cluster sensor readings than K-Means can do. Finally, a classification model is built to classify unseen sensor readings and identify the road surface quality. An accuracy of 96% can be obtained from the built classifier confirming the effectiveness of the adopted methodology in evaluating the road surface quality in Egypt.