Recent computer vision applications for pavement distress and condition assessment

Amidst the unprecedented demographic boom, coupled with climate change, more pressure is being exerted on road networks. Asset managers are thus in search for time and cost-effective state-of-the-art technologies for road inspection and condition monitoring. This paper provides an up-to-date comprehensive review of Computer Vision (CV) models and applications in pavement distress detection, classification, segmentation, quantification and condition assessment. To this end, the objectives of this review are: (1) review and bibliometric analysis of 190 related recent publications; (2) identification of trending tools, research gaps, emerging technologies, challenges and limitations of using CV for pavement distress and condition assessment; and (3) guiding of future research related to CV pavement asset management. While CV related models saw a sharp increase recently, increased collaboration between the academia and the industry is still needed to improve applicability levels of such models by pavement management agencies.