A Transfer Learning Methodology for Recognizing Unsafe Behavior during Lifting Operations in a Chemical Plant

Large lifting equipment is used regularly in the maintenance operations of chemical plant installations, where safety controls must be carried out to ensure operations. This paper presents a convolutional neural network (CNN) methodology, based on PyTorch framework, identify unsafe behavior among operation drivers, specifically, by collecting 22,352 images over certain time period plant. The drivers’ was divided into eight categories, and ResNet50 model selected pictures. results show that proposed transfer learning achieves 99.6% accuracy rate, 99% recall rate F1 value for expected behaviors drivers. knowledge regarding industry provides new perspective preventing accidents caused dangerous