Distracted driving behavior has become a leading cause of vehicle crashes. This paper proposes a data augmentation method for distracted driving detection based on the driving operation area. First, the class activation mapping method is used to show the key feature areas of driving behavior analysis, and then the driving operation areas are detected by the faster R-CNN detection model for data augmentation. Finally, the convolutional neural network classification mode is implemented and evaluated to detect the original dataset and the driving operation area dataset. The classification result achieves a 96.97% accuracy using the distracted driving dataset. The results show the necessity of driving operation area extraction in the preprocessing stage, which can effectively remove the redundant information in the images to get a higher classification accuracy rate. The method of this research can be used to detect drivers in actual application scenarios to identify dangerous driving behaviors, which helps to give early warning of unsafe driving behaviors and avoid accidents.

Distracted driving is the leading cause of road traffic accidents. It is essential to monitor the driver's status to avoid traffic accidents caused by distracted driving. Current research on detecting distracting behaviours focuses on analysing image features using convolutional neural networks (CNNs). However, the generalisation ability of the current distracted driving models is limited. This paper aims to improve the generalisation ability of distracted driving models that are affected by factors such as the driver himself, the background, the monitoring angle, and so on. A new driver distraction detection method, which is referred to as multi‐scale domain adaptation network (MSDAN), was proposed to improve model adaptability. The method consists of three stages: first, multi‐scale convolution was introduced to build a new backbone to accommodate better the valuable feature of the target on different scales. Secondly, the authors designed the domain adaptation network to improve the model's adaptability to the difference in data sources through adversarial training. Finally, dropout is added to the fully connected layer to increase the model's generalisation ability. The comparison results on the large‐scale driver distraction detection dataset show that the authors’ method can accurately detect driver distraction and has good generalisation performance, with an accuracy improvement in the cross‐driver and cross‐dataset experiments.

Detecting distracted driving accurately and quickly with limited resources is an essential yet underexplored problem. Most of the existing works ignore the resource-limited reality. In this work, we aim to achieve accurate and fast distracted driver detection in the context of embedded devices where only limited memory and computing resources are available. Specifically, we propose a novel convolutional neural network (CNN) light-weighting method via adjusting block layers and shrinking network channels without compromising the model’s accuracy. Finally, the model is deployed on multiple devices with real-time detection of driving behaviour. The experimental results for the American University in Cairo (AUC) and StateFarm datasets demonstrate the effectiveness of the proposed method. For instance, for the AUC dataset, the proposed MobileNetV2-tiny model achieves 1.63% higher accuracy with just 78% of the model parameters of the original MobileNetV2 model. The inference speed of the proposed MobileNetV2-tiny model on resource-limited devices is on average 1.5 times that of the original MobileNetV2 model, which can meet real-time requirements.