AB-DLM: An Improved Deep Learning Model Based on Attention Mechanism and BiFPN for Driver Distraction Behavior Detection

Abstract

Driver distraction behavior causes a large number of traffic accidents every year, resulting in economic losses and injuries. Currently, the driver still plays an important role in the driving and control of the vehicle due to the low level of vehicle automation and the immature development of autonomous driving. Therefore, it is vital to research distraction detection for drivers. However, in realistic driving scenarios with uncertain information, they are still some challenges in efficient and accurate driver distraction detection. In this paper, an improved deep learning model based on attention mechanisms and bi-directional feature pyramid networks (BiFPN) is proposed to identify driver distractions. Firstly, an improved data augmentation strategy is introduced to increase the data diversity to enhance the generalization capability of the model. Secondly, the squeeze-and-excitation (SE) attention mechanism layer is used after the C3 module of the original backbone network to enhance the important feature information and suppress the minor feature information. Finally, the BiFPN module is introduced into the neck network to better achieve multi-scale feature fusion without increasing the calculation amount too much. The experimental results show that the method proposed in this paper has an average mean accuracy rate (mAP) of 0.956 on the test set. Compared to the original model the mAP has improved by 13.2%. The detection speed of the model is 71 frames per second, and the memory occupation is 15.9 MB. This method has the advantages of high recognition accuracy, fast detection speed, and small memory occupation of the model, which are important for achieving engineering deployment.

Swin-Transformer -YOLOv5 for lightweight hot-rolled steel strips surface defect detection algorithm

“…The feature fusion method of YOLOv5 using PANet is changed in this paper to feature fusion using BiFPN to construct a feature pyramid, and the semantic features extracted from the backbone network are fused top-down using efficient bi-directional cross-scale connectivity and weighted feature map fusion. The shallow network can contain clearer location information due to larger resolution, and the deep network can contain more high-dimensional semantic information due to the large sensory field [35,36]. More features can be fused without increasing the cost by adding lateral connections between the original input and output nodes of the same feature.…”

Defect detection of solar panels using improved faster R-CNN

“…The strong semantic information of the top layer feature map is transmitted downward to obtain the feature map with strong semantic information and high resolution, and reduce the false check rate of the lower layer feature map. The improved feature fusion structure optimizes the feature fusion into weighted feature fusion [8], distinguishes the importance of each weight, and uses the normalization fusion calculation. Take 2 p fusion process as an example,the parameters and calculation process are shown in Figure 4.…”

Optimizing Road Safety: Advancements in Lightweight YOLOv8 Models and GhostC2f Design for Real-Time Distracted Driving Detection

“…While these methods achieve balance between accuracy and speed, they fall short in enhancing detection for diverse target scales. Li et al [24] merged attention mechanisms with bidirectional feature pyramid networks, boosting the model's capacity to amalgamate diverse-scale object data. This led to a remarkable 95.6% accuracy on the SFDDD dataset, albeit accompanied by an extra computational load.…”