Automated assessment of bridge guardrails for regional prioritization based on open-source data and deep learning algorithms

• The compliance of bridge safety barriers is an often-overlooked issue. • A novel approach combining open-source data and deep learning is proposed. • The automated barrier classification reduces the need for time-consuming inspections. • OpenStreetMap and Google Street View are used to build bridge visual datasets. • Imagery is interpreted by YOLO deep learning model to assess barrier compliance. Guardrails installed on road bridges are essential to the safety of vehicles and road users. However, infrastructure managers often lack accurate and up-to-date inventories of the safety barriers on their bridges. Moreover, the limited availability of financial resources, combined with the lack of adequately trained personnel, makes it difficult for these entities to effectively assess the condition of such critical infrastructure, even though, in most cases, it is only necessary to recognize the barrier code compliance, which could be done even by less trained personnel. This paper proposes an integrated methodology to automate the assessment of road guardrails installed on bridges using open-source data and deep learning (DL) algorithms. Besides the use of the consolidated YOLO (You Only Look Once) object detection algorithm to classify the safety barriers to establish whether they match the current standards, the process innovatively involves the extraction of bridge information from OpenStreetMap (OSM) to construct a database of existing bridges. This latter step is integrated with Google Street View API, for the extraction of images of each bridge’s safety barriers to be analysed by YOLO. The synergic concatenation of these three steps (OSM, Google Street View, YOLO) into a unique software tool, provides road managers with a cost-effective and efficient tool to remotely survey the guardrails installed on their bridges, permitting to prioritize maintenance and upgrading interventions. The methodology not only facilitates the task of infrastructure monitoring but also helps ensure the safety of road users by timely identifying non-compliant safety barriers. Finally, the developed procedure allowed to identify non-conforming guardrails on a subset of bridges located in a southern Italy region permitting to compute the intervention cost for their replacement.