Fatigue detection method for UAV operators based on depth perception technology

The study proposes a real-time fatigue detection method based on deep perception technology, specifically designed for Unmanned Aerial Vehicle (UAV)operators to ensure safe operation. The approach leverages Kinect 2.0 to capture the facial and posture information of the operator. Feature extraction is performed using a Convolutional Neural Network, in combination with the percentage of eyelid closure rate over time (PERCLOS) algorithm, to accurately identify fatigue states. Experimental results demonstrate that this method achieves an accuracy rate of 93.3% in fatigue detection, outperforming traditional methods. Future work will focus on further optimizing the model structure and integrating multimodal data to enhance the system’s precision and adaptability, providing technological support for UAV safety management.