Fig. 5
From: YOLO-DP: A detection model of fifteen common rice diseases and pests
Overview of the global-to-local spatial aggregation module (GLSA). This figure shows the module consists of two parallel layers: a GSA (Global Self-Attention) layer, which focuses on the content of the pixels based solely on their content and a GLA (Global Location Attention) layer, which focuses on the spatial location of the pixels. The output of the module is the sum of the outputs from these two layers.
