Truck Driver Fatigue Detection Based on Video Sequences in Open-Pit Mines

Due to complex background interference and weak space–time connection, traditional driver fatigue detection methods perform poorly for open-pit truck drivers. For these issues, this paper presents a method based on Libfacedetection an LRCN. The consists of three stages: (1) using face module with tracking quickly extract the ROI face; (2) extracting coding features; (3) combining model build spatiotemporal classification network. innovation is utilize features image sequence suitable task. Meanwhile, added stage reduce time expenditure. As result, average speed video increased by 74% in comparison one without method. Our best adopts DHLSTM feature-level frame aggregation, which achieves high accuracy 99.30% self-built dataset.