Fig. 1

YOLOv8n architecture diagram. The YOLOv8n architecture consists of three main components: the Backbone, Neck, and Head. The Backbone extracts features from the input image using a series of convolutional and C2f modules, followed by a Spatial Pyramid Pooling Fast (SPPF) block for multi-scale feature aggregation. The Neck adopts a feature fusion strategy with upsampling, concatenation, and additional C2f layers to enhance multi-level representation. The Head performs prediction at multiple scales and outputs object classification and bounding box regression. Supplementary diagrams for SPPF and C2f modules are shown at the bottom for structural clarity.