Object Detection Based on YOLOv5 and GhostNet for Orchard Pests

Real-time detection and identification of orchard pests is related to the economy industry. Using lab picture collections pictures from web crawling, a dataset common in orchards has been created. It contains 24,748 color images covers seven types pests. Based on this dataset, paper combines YOLOv5 GhostNet explains benefits method using feature maps, heatmaps loss curve. The results show that mAP proposed increases by 1.5% compared original YOLOv5, with 2× or 3× fewer parameters, less GFLOPs same time. Considering parameters Ghost convolution, our new can reach higher epochs. Smaller neural networks are more feasible deploy FPGAs other embedding devices which have limited memory. This research provides algorithm devices.