Fig. 3
From: YOLO-DP: A detection model of fifteen common rice diseases and pests
Structure diagram of YOLO-DP. This figure illustrates the structure of YOLO-DP, including the Backbone, the Neck, the Head, the Loss, the Input and the Output. The backbone consists of Conv: A standard convolutional layer for extracting features, WTConv: Improved convolutional layers, GLSA: the Global to Local Spatial Aggregation) module, Concat: Feature stitching operation to fuse features at different levels, C2f: Feature Processing Module, Detect: The detection head is used to generate target detection results, and EloU:Loss function. The input and output image size is 640 × 640 × 3.
