Vision-based real-time process monitoring and problem feedback for productivity-oriented analysis in off-site construction

Abstract

The widespread adoption of surveillance cameras in work environments has enabled the direct and non-intrusive detection of productivity-related issues in the field of construction. In this research, a process monitoring and problem feedback framework is developed based on closed-circuit television footage and computer vision analysis to achieve real-time visual control of the work process and facilitate data-driven decision-making in off-site construction. To enhance the automation of productivity-related problem recognition, a novel video analysis algorithm is developed to process the inputted video footage and provide feedback with respect to nine productivity issues. The z-score and Exponential Moving Average methods are employed to eliminate detection errors, and the spatial density analysis method is adopted to visually analyze spatial information. The observed performance of the proposed framework demonstrates that it can accurately acquire data from footage and provide process monitoring and problem feedback in real time.