Fig. 1: Experimental workflow for mobile-based road monitoring.
From: Lightweight deep learning for real-time road distress detection on mobile devices
Overview of data processing, model optimization, mobile deployment, and on-site detection. a data collection and augmentation. Icon by Leonardo Henrique Martini from The Noun Project (CC BY 3.0). b MobiLiteNet Framework for Lightweight Model Deployment. Icon by HideMaru from The Noun Project (CC BY 3.0). c Server-to-android smartphone and mixed reality device. Contains elements by juicy_fish and macrovector via Freepik Free License. d On-site detection. Map tiles adapted from Stamen Design under CC BY 3.0. Data © OpenStreetMap contributors (ODbL). e Expansion of public database. Icons by Tini Sumiarsih and Alzam from The Noun Project (CC BY 3.0).
