Abnormal Cockpit Pilot Driving Behavior Detection Using YOLOv4 Fused Attention Mechanism

Abstract

The abnormal behavior of cockpit pilots during the manipulation process is an important incentive for flight safety, but the complex cockpit environment limits the detection accuracy, with problems such as false detection, missed detection, and insufficient feature extraction capability. This article proposes a method of abnormal pilot driving behavior detection based on the improved YOLOv4 deep learning algorithm and by integrating an attention mechanism. Firstly, the semantic image features are extracted by running the deep neural network structure to complete the image and video recognition of pilot driving behavior. Secondly, the CBAM attention mechanism is introduced into the neural network to solve the problem of gradient disappearance during training. The CBAM mechanism includes both channel and spatial attention processes, meaning the feature extraction capability of the network can be improved. Finally, the features are extracted through the convolutional neural network to monitor the abnormal driving behavior of pilots and for example verification. The conclusion shows that the deep learning algorithm based on the improved YOLOv4 method is practical and feasible for the monitoring of the abnormal driving behavior of pilots during the flight maneuvering phase. The experimental results show that the improved YOLOv4 recognition rate is significantly higher than the unimproved algorithm, and the calling phase has a mAP of 87.35%, an accuracy of 75.76%, and a recall of 87.36%. The smoking phase has a mAP of 87.35%, an accuracy of 85.54%, and a recall of 85.54%. The conclusion shows that the deep learning algorithm based on the improved YOLOv4 method is practical and feasible for the monitoring of the abnormal driving behavior of pilots in the flight maneuvering phase. This method can quickly and accurately identify the abnormal behavior of pilots, providing an important theoretical reference for abnormal behavior detection and risk management.

YOLOv5s-MEE: A YOLOv5-based Algorithm for Abnormal Behavior Detection in Central Control Room

“…For example, Mo et al [20] proposed a research on human behavior detection based on Faster R-CNN. To increase the detection accuracy for small target, Chen et al [3] presented an abnormal driving behav-ior (including smoking and calling) detection algorithm based on YOLOv4, the research introduces the CMBA attention mechanism to enhance the detection of small target. The proposed model shows good performance on self-constructed dataset.…”

Advanced Machine Learning Applications in Big Data Analytics

“…Chen et al [22] proposed a method for detecting abnormal pilot behavior during flight based on an improved YOLOv4 deep learning algorithm and an attention mechanism. The CBAM attention mechanism was introduced to improve the feature extraction capability of the deep neural network.…”

An improved GhostNet for unsafe driving behavior algorithm

“…Majdi [4] proposed a Drive-Net model to identify driving behavior and added a random forest algorithm after the fully connected layer to get the final result. Song [1] and Chen [5] respectively proposed a driving behavior detection algorithm based on object detection, but can only identify two or three kinds of unsafe driving behaviors.…”