Research on Pedestrian Target Detection Method Based on Improved YOLOv8 Algorithm

In the field of target detection, particularly pedestrian detection in complex scenarios, accuracy and real-time performance are persistent challenges. To address this issue, this paper introduces an enhanced model named MAM-Guided YOLOv8, which builds upon YOLOv8 by incorporating four distinct attention mechanisms: Local Attention Mechanism (LAM), Global Attention Mechanism (GAM), Efficient Channel Attention (ECA), and Axial Attention Mechanism (AA). Through a meticulously designed attention mechanism fusion strategy, the MAM-Guided YOLOv8 model aims to improve pedestrian detection capabilities in complex environments. Experimental results demonstrate that this model outperforms traditional methods on urban complex scene datasets, effectively balancing high precision with rapid processing demands. The contribution of this paper lies in the innovative integration of multiple attention mechanisms, significantly enhancing pedestrian detection accuracy and providing an effective solution for object detection tasks in practical applications.