A Bibliometric analysis of computer vision applications in construction and demolition waste management

Abstract

With approximately 10 billion tonnes of construction and demolition waste produced yearly, the construction sector is the biggest producer of solid waste, which accounts for 40% of urban solid waste that contributes to serious environmental issues, including soil degradation, resource depletion, and an excessive dependence on landfilling. Introduction of new avenues towards addressing these issues has been facilitated by recent advancements in computer vision that have yielded efficient, self-sustaining, and prudent construction and demolition waste management. The objective of this study is to conduct a rigorous bibliometric analysis to examine the role of computer vision towards improved construction and demolition waste management. This study will identify the growth of the research area, top contributing sources, and collaboration patterns. A bibliometric approach was employed to evaluate 29 peer-reviewed journal articles and conference proceedings between 2015 and 2025. Organized keyword searching to obtain information from Scopus and Web of Science was employed. Manual analysis was done for identifying publication patterns, top sources, and country output. Keyword co-occurrence and collaboration network analysis were done using VOS viewer software. Although computer vision techniques, including semantic segmentation, object detection, and deep learning, have been promising in recycling and waste sorting automation, existing literature is fragmented. Most studies are focused on a single application, without the extraction of broader trends or collaborative dynamics. The paper provides an integrated corpus of knowledge for researchers, practitioners, and policymakers by presenting an exhaustive bibliometric analysis that documents the historical patterns of computer vision in construction and demolition waste research.