Computer vision-based construction progress monitoring

• Published in: Automation in construction Vol. 138; p. 104245
• Main Authors: Reja, Varun Kumar, Varghese, Koshy, Ha, Quang Phuc
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.06.2022; Elsevier BV
• Subjects: 3D reconstruction; Algorithms; As-built modelling; Automated construction; Computer vision; Construction; Data acquisition; Digital Twin; Image reconstruction; Literature review; Monitoring; Point cloud; Progress monitoring; Scan to BIM; Three dimensional models; CV-CPM; LoD; Scan to BIM; BoQ; UGV; As-built modelling; Literature review; DL; GPS; GPU; DT; RGBD; 3D reconstruction; GUI; Progress monitoring; AEC; HSI; ML; Automated construction; Computer vision; QR; Data acquisition; MR; Digital Twin; CAD; AI; PSR; PCA; IMU; AR; Point cloud; CMVS; SMS; MEP; SfM; VR; CNN; IFC; LPM; MVS; UAV; IoT; TLS; PMVS; SLAM; BIM; LiDAR; RFID; RANSAC; ICP; SGM; XR; CCTV
• ISSN: 0926-5805; 1872-7891
• DOI: 10.1016/j.autcon.2022.104245

Automating the process of construction progress monitoring through computer vision can enable effective control of projects. Systematic classification of available methods and technologies is necessary to structure this complex, multi-stage process. Using the PRISMA framework, relevant studies in the area were identified. The various concepts, tools, technologies, and algorithms reported by these studies were iteratively categorised, developing an integrated process framework for Computer-Vision-Based Construction Progress Monitoring (CV-CPM). This framework comprises: data acquisition and 3D-reconstruction, as-built modelling, and progress assessment. Each stage is discussed in detail, positioning key studies, and concurrently comparing the methods used therein. The four levels of progress monitoring are defined and found to strongly influence all stages of the framework. The need for benchmarking CV-CPM pipelines and components are discussed, and potential research questions within each stage are identified. The relevance of CV-CPM to support emerging areas such as Digital Twin is also discussed. [Display omitted] •A comprehensive review and categorisation of recent studies on computer vision-based construction progress monitoring.•An integrated framework for Computer Vision-Based Construction Progress Monitoring (CV-CPM).•Establishing a need for exploring a hybrid approach (heuristic-learning based) for as-built modelling.•Categorisation of on-site progress monitoring requirements into four levels and associating these with the CV-CPM framework.•Discussion on usage of the proposed framework for developing a strategy and roadmap to enable benchmarking in CV-CPM.