Driver distraction detection based on adaptive tiny targets and lightweight networks

Driver distraction detection is critical to reducing road traffic accidents and increasing the efficiency of advanced driver assistance systems. Real-time lightweight models are especially important for in-vehicle devices with limited computing resources. However, most existing methods focus on designing lighter network architectures and ignore the performance loss when detecting tiny targets. In order to realize the collaborative optimization of tiny target detection accuracy and network lightweight, a driver distraction detection method ATD 2 Net based on adaptive tiny target detection and lightweight networks is proposed. This method aims to reduce model complexity while fully capturing target features for accurate detection. ATD 2 Net consists of three core modules, Channel Reconstruction Perception Module (CRPM), Dynamic Spatial Self-locking Module (DSSM) and Structural Feedback Optimization Module (SFOM). CRPM reconfigures channels and reconstructs them into batch dimensions, uses parallel strategies to perceive interactive features between channels, and significantly enhances feature extraction capabilities. DSSM adopts dynamic locking and adaptive spatial selection mechanisms to capture multi-scale features while injecting adaptive spatial information. It effectively aggregates instance features and reduces the interference of conflicting information and background information, thereby improving the detection ability of tiny targets. SFOM uses dependency trees to model inter-layer relationships and integrate coupling parameters into groupings. It uses a sparse strategy to remove unimportant parameters, achieving lightweight modeling while balancing accuracy and speed. Experimental results show that ATD 2 Net is superior to the latest methods in driver distraction detection, showing excellent performance and good application prospects. • A CRPM module (Channel Reconstruction Perception Module) is proposed to effectively configure channel sizes through layer index parameterization and parallel strategies, capturing inter-channel dependencies while reducing computational overhead. • A DSSM module (Dynamic Spatial Self-locking Module) is proposed to extract multi-scale target information by dynamically locking and adjusting the background focus range and fuse the core tiny target information through adaptive spatial selection while preventing feature disappearance. • A SFOM module (Structural Feedback Optimization Module) is proposed to model inter-layer and intra-layer relationships through dependency trees, assign group-level weights, perform sparse training, and iteratively prune redundant weights through feedback to achieve structural re-adjustment.