Research on Crop Pest Detection Based on Improved YOLO11n Model

Abstract

With the continuous development of agricultural intelligent technology in China, crop pest detection has become an important part of ensuring crop yield and quality, gradually emerging as a research hotspot and challenge. To achieve more accurate and efficient identification of crop pests, this paper proposes an intelligent detection algorithm for crop pests based on the improved YOLO11n model. The CBAM attention mechanism is introduced into both the Neck and Backbone structures of YOLO11n to enhance feature extraction and fusion, assign greater weights to important targets, and improve object detection accuracy. Experimental verification shows that after introducing the CBAM attention mechanism into the Neck and Backbone structures, Precision, Recall, mAP@0.5, and mAP@0.5:0.95 all increase, with mAP@0.5 improving by 4.29 %. This proves that the CBAM attention mechanism is suitable for these positions and can enhance model performance. This study completes the improvement of the YOLO11n model, achieving more accurate and efficient pest identification.