Automated dimensional quality assessment for formwork and rebar of reinforced concrete components using 3D point cloud data

Abstract

Dimensional quality assessment (DQA) on the formwork and rebar of a reinforced concrete (RC) element is an important task for factory or field inspectors to ensure compliance with the blueprints prior to casting concrete. Current practice for the formwork and rebar DQA of RC elements relies on measurement-tape based manual inspection by qualified personnel, which is labor intensive and time consuming. In addition, the current practice for the inspection often involve the case that field inspectors need to walk onto rebar cages for close assessments, leading to potential safety hazard and damage the integrity of the structure. In order to tackle these limitations of the current practice and research, this study aims to develop a laser scanning-based technique that automatically assesses the key DQA checklists of RC element including rebar spacing and concrete cover with respect to the formwork. To this end, a noise removal algorithm is developed based on the known geometric configuration of formwork and rebar to remove background noise and mixed pixels. Key features of the formwork and rebar are then automatically extracted using the principal component analysis and the RANSAC. The experimental results show that the rebar spacing, the formwork dimension, the concrete cover and the side cover of the tested specimen are estimated with discrepancies of 2.15 mm, 2.52 mm, 2.18 mm and 3.12 mm on average, respectively, demonstrating that the proposed technique can provide accurate solution for the formwork and rebar DQA during the fabrication stage before pouring concrete. In addition, the lessons learned from the results of the laboratory test including the issues of scan density, formwork material are comprehensively discussed, which will be useful for practitioners.