YOLOv9 Algorithm Improvement for Small Object Detection in UAV Aerial Imagery

This paper addresses the challenges of small object detection in UAV aerial imagery by proposing an improved YOLOv9 object detection model.The core innovations are twofold: first, the introduction of a Global Attention Mechanism (GAM), which enhances the model's perception of small object features through channel and spatial dual-path attention processing; second, the adoption of a Bidirectional Feature Pyramid Network (BiFPN), which implements bidirectional feature fusion from top-down and bottom-up perspectives, effectively improving the interaction efficiency of features at different scales.Experiments on the VisDrone2019 dataset demonstrate that, compared to the baseline YOLOv9, the proposed model improves the [email protected] metric by 1.6%, while reducing the parameter count by 2.810^6.Visual comparisons show that the model exhibits superior detection capabilities in complex environments such as aerial multi-scale small objects, dense crowds, and night scenes, effectively addressing the problems of missed detections and false detections of small objects.