Fig 1.
Structural comparison diagram of the C3 module and the C2f module.
Fig 2.
MGAC-YOLO structure diagram.
Fig 3.
MConv structure diagram.
Fig 4.
GAM structure diagram.
Fig 5.
CloAttention structure diagram.
Fig 6.
GACAttention structure diagram.
Fig 7.
Structure diagram of the improved feature extraction.
Table 1.
Training parameter configuration.
Table 2.
Results of ablation experiments.
Fig 8.
Visual comparison chart of training results.
Table 3.
Results of comparative experiments.
Fig 9.
Detection result of MGAC-YOLO in dark scenes.
Fig 10.
Detection result of M-YOGACLOv8 under conditions of shadows and dynamic blur.
Fig 11.
Detection result of MGAC-YOLOv8 in occluded scenes.
Fig 12.
Detection result of MGAC-YOLOv8 in high-altitude photography.
Fig 13.
Visual comparative experiment.