Vision-Based Pavement Marking Detection and Condition Assessment—A Case Study

Pavement markings constitute an effective way of conveying regulations and guidance to drivers. They the most fundamental communicate with road users, thus, greatly contributing ensuring safety order on roads. However, due increasingly extensive traffic demand, pavement are subject a series deterioration issues (e.g., wear tear). Markings in poor condition typically manifest as being blurred or even missing certain places. The need for proper maintenance strategies roadway markings, such repainting, can only be determined based comprehensive understanding their as-is worn condition. Given fact that efficient, automated accurate approach collect information is lacking practice, this study proposes vision-based framework marking detection assessment. A hybrid feature detector threshold-based method were used line identification classification. For each identified marking, its worn/blurred severity level was then quantified terms percentage at pixel level. damage estimation results compared manual measurements evaluation, indicating proposed capable providing indicative knowledge about markings. This paper demonstrates promising potential computer vision infrastructure sector, implementing wider range managerial operations management.