No abstract

No abstract

No abstract

Automated worker action recognition helps to understand the states of workers, enabling effective management of work performance in terms of productivity, safety, and health issues. A wristband equipped with an accelerometer (e.g., activity tracker) allows us to collect the data related to workers' hand activities without interfering with their ongoing work. Considering that many construction activities involve unique hand movements, the use of acceleration data from a wristband has great potential for action recognition of construction activities. In this context, the authors examine the feasibility of the wrist-worn accelerometer embedded activity tracker for automated action recognition. Specifically, masonry work was conducted to collect acceleration data in a laboratory. The classification accuracy of four classifiers, such as the k-nearest neighbor, multi-layer perceptron, decision tree, and multi-class support vector machine, was analyzed with different window sizes to investigate classification performance, and it was found that the multiclass support vector machine with a 4-second window size showed the best accuracy (88.1%) to classify 4 different sub-tasks of masonry work. The present study makes one noteworthy

No abstract