|
| Abstract: |
| Real-time detection of distracted driving behavior by drivers can effectively prevent traffic accidents from occurring. This study presents a lightweight distracted driving behavior detection approach leveraging the RT-DETR-R18 (real-time detection transformer with ResNet-18) approach to mitigate the real-time performance limitations commonly observed in existing distracted driving detection methods. To address the difficulty of maintaining model accuracy after lightweight improvement, a dual backbone dynamic feature fusion network (DB-DFN) is proposed to replace ResNet-18 for early feature extraction and fusion. An additive-based feature fusion module (AddMixer) is introduced to replace attention-based intra-scale feature interaction (AIFI) in the original model for efficient feature interaction and to improve testing speed. The convolution module with an integrated gating mechanism was replaced with the original model's convolution module to improve cnn-based cross-scale feature fusion(CCFM), reducing redundant computation of the model's convolutional layer. This study conducted experiments leveraging the State Farm Distracted Driver Detection dataset. Compared with the RT-DETR-R18 model, the proposed method minimizes the number of parameters by 34.8%, lowers the computational cost by 49.8%, and increases the inference speed by 34.8 frames per second(FPS). Meanwhile, the mAP50 remains unchanged, indicating that detection accuracy is preserved while real-time performance is significantly enhanced through lightweight optimization. |
| Key words: RT-DETR-R18 distracted driving behavior detection DB-DFN cross-scale feature fusion |
| DOI:10.11916/j.issn.1005-9113.25013 |
| Clc Number:TP183 |