AI-Based Concrete Compaction Completion Judgment Technology
By T. Hayashi, R. Takagi, H. Kojima and S. Nagata

Concrete Journal, Vol.60, No.3, Mar. 2022


Synopsis
The compacting of concrete is an important work process that affects the performance of concrete after hardening. Traditionally, the degree of compaction of concrete is evaluated visually based on the changes in the condition of the concrete surface, and this evaluation is largely dependent on the experience of the technician performing the compaction work. As an alternative to visual evaluation by technicians, the authors have devised and verified an AI-based judgment method. They confirmed that AI can judge the completion of compaction in real time in a manner similar to that of technicians by assigning technicians' judgments as correct answer labels to the frame images obtained from the concrete compaction video and by taking into account numerical data indicating the properties of the concrete in the fresh state, such as slump and air content.
Keywords:
Compacting, Completion judgment, Visual evaluation, AI, Deep learning, Productivity improvement

To Pagetop