YOLOV8-KF: A FRAMEWORK FOR PEST AND DISEASE DETECTION AND FRUIT-VEGETABLE RECOGNITION IN COMPLEX SCENARIOS
Keywords:
Crop disease, Object detection, YOLOv8Abstract
Crop pests and diseases constitute major constraints affecting crop yield and quality, and pose a threat to global food security. Therefore, in modern agricultural production, accurate and real-time detection of pests and diseases is critical for the timely implementation of prevention and control measures. Although YOLOv8 performs excellently in multi-class pest and disease detection as well as fruit and vegetable recognition, it yields unsatisfactory detection results under scenarios involving leaf occlusion and complex field monitoring.To address these issues, this study develops a framework for pest and disease detection and fruit and vegetable recognition that combines YOLOv8 with Kalman filter post-processing. The Kalman filter is used to predict target positions and smooth detection deviations under occluded conditions, thereby improving the continuity and stability of target tracking.The results show that the proposed model not only maintains high recognition accuracy for small-sized and multi-class pests, diseases and fruit-vegetable targets, but also reduces missed detection caused by occlusion. This method provides an effective solution for real-time dynamic monitoring of crop pests and diseases, fruit and vegetable recognition, and precise prevention and control in complex scenarios.References
[1] Zhang Shugui, Chen Shuli, Zhao Zhan. Improved YOLOv8 algorithm for identifying crop leaf diseases and pests. Chinese Journal of Agricultural Mechanization, 2024, 45(07): 255-260. DOI: 10.13733/j.jcam.issn.2095-5553.2024.07.038.
[2] Lee J, Yin H, Jin D, et al. PestDetectSim: an integrated approach for crop pest diagnosis using object detection and similarity-based image retrieval. Plant methods, 2026. DOI: 10.1186/S13007-026-01520-X.
[3] Shi Jie, Xiong Kaixiang, Li Zhi, et al. A maize pest and disease detection system based on a lightweight improved YOLOv8 model and edge computing. Jiangsu Journal of Agricultural Sciences, 2025, 41(02): 313-322.
[4] Yu X, Zhang H, Duan Y, et al. An optimized convolutional neural network based on multi-strategy grey wolf optimizer to identify crop diseases and pests. Displays, 2026, 92: 103341. DOI: 10.1016/J.DISPLA.2026.103341.
[5] Xiong Xianhua, Yang Zheng, Cheng Siqin, et al. A method for ship image recognition and tracking in lock chambers based on improved YOLOv8 and Kalman filter algorithm. China Harbour Construction, 2025, 45(06): 17-24.
[6] Wang Yimeng, Jin Xiaoke. Research on Intelligent Detection of Crop Diseases and Pests and Precision Spraying by Agricultural Machinery Based on Image Processing. China Agricultural Machinery Equipment, 2026(03): 55-57.
[7] Gao Q, Shi C, Ji Y, et al. Crop Pest Identification and Real-Time Monitoring System Design Based on Improved YOLOv8s. Sensors, 2026, 26(2): 404. DOI: 10.3390/S26020404.
[8] Fang Yingying, Pan Yanhong, Pan Xinling, et al. Design of a deep learning-based crop disease and pest identification system. Smart Agriculture Guide, 2026, 6(03): 23-28. DOI: 10.20028/j.zhnydk.2026.03.006.
[9] Huang R, Li M, Zheng H, et al.Chinese crop diseases and pests named entity recognition based on variational information bottleneck and feature enhancement. Scientific Reports, 2025, 15(1): 31573. DOI: 10.1038/S41598-025-04252-5.