Road Anomaly Detection Techniques: A Systematic Review for Enhancing Vehicle and Driver Safety

Road anomalies like potholes, cracks, and debris pose significant risks to vehicle safety and driver comfort, increasing accident rates and maintenance costs. As urbanization and vehicular traffic grow, traditional road monitoring methods struggle to meet the need for real-time, scalable, and accurate anomaly detection. Advances in technology have shifted research in intelligent transportation systems toward the automated detection of road anomalies. This review examines state-of-the-art methods from repositories such as Google Scholar, Scopus, IEEE, Science Direct, MDPI, DOAJ, and Hindawi over the past eight years, covering traditional image processing to advanced machine learning algorithms and deep learning models. By analyzing existing literature, it identifies challenges like the need for robust datasets, real-time processing, and model generalization. The paper also addresses research gaps and suggests future directions for scalable road anomaly recognition systems. This review aims to support researchers, academicians, and tech enthusiasts in understanding the technologies for detecting road anomalies, contributing to safer roads and smoother driving experiences for both human drivers and autonomous vehicles.