MAFIKD: A Real-Time Pest Detection Method Based on Knowledge Distillation

The significant damage caused by pests to crops has always been a pressing issue in agricultural production. To address the problems of low recognition accuracy, weak feature extraction capability, and poor robustness of lightweight pest detection models, this study proposes a knowledge distillation (KD) algorithm MAFIKD based on multiattention feature fusion (MA) and adaptive fine-grained feature imitation (FI). MAFIKD combines MA and FI to enhance the attention of the student to the key features of the teacher, establishing diversified knowledge such as feature correlation and sample correlation to alleviate the difficulty of knowledge transfer in pest detection models. Based on you only look once version 5 (hereinafter referred to as YOLOv5), we used a self-made pest dataset to evaluate the proposed algorithm. Experimental results show that after applying MAFIKD, YOLOv5-CSPDarknet achieved 85.7% mean average precision (mAP)@0.5 and 76.12% mmAP, which are 3.13% and 4.56% higher than the baseline, respectively. To verify the actual inference speed of the model, this study developed a mobile application (APP) for pest detection based on Android, using the ncnn convolutional neural network (NCNN) high-performance neural network forward computing framework to deploy the pest detection model offline to mobile terminals, and deployed the model on the server using the Nginx+uWSGI+Flask architecture to provide online and offline pest detection services. The experimental results show that after applying MAFIKD, YOLOv5-CSPDarknet achieved an average detection frame rate of 10.1 frames per second (FPS) on the HUAWEI Enjoy 20, and the model size was only 14.5 MB, meeting the real-time detection requirements for field pests.