轻量级深度学习网络在农作物目标检测的应用进展  

Review on lightweight deep learning networks for object detection in crops

作　　者：许毓超 吴茜 张兵园 周玲莉 任妮 张美娜[1,2,3] Xu Yuchao;Wu Qian;Zhang Bingyuan;Zhou Lingli;Ren Ni;Zhang Meina(School of Agricultural Engineering,Jiangsu University,Zhenjiang,212013,China;Institute of Agricultural Information,Jiangsu Academy of Agricultural Sciences/Key Laboratory of Intelligent Agricultural Technology(Yangtze River Delta),Ministry of Agriculture and Rural Affairs,Nanjing,210014,China;Zhongshan Biological Breeding Laboratory,ZSBBL,Nanjing,210014,China)

机构地区：[1]江苏大学农业工程学院,江苏镇江212013 [2]江苏省农业科学院农业信息研究所/农业农村部长三角智慧农业技术重点实验室,南京市210014 [3]生物育种钟山实验室,南京市210014

出　　处：《中国农机化学报》2025年第3期261-270,共10页Journal of Chinese Agricultural Mechanization

基　　金：国家自然科学基金项目(32201664);江苏省重点研发计划(BE2022363);江苏省农业科技自主创新资金项目(CX(22)5009);江苏省创新能力建设计划(BM2022008—01)。

摘　　要：随着计算机视觉领域中深度学习网络模型应用的发展,各类农业场景中的目标检测性能得到极大的推动。与部署在云端服务器的大规模深度学习网络不同,轻量级深度学习网络因其较小的参数量和运算量,在硬件资源有限且实时性要求更高的农业场景中展现出潜力,完成果蔬采摘机器人的目标检测、作物病虫草害目标检测以及作物表型检测等任务。概述当前主流轻量级深度学习网络的模型结构、关键技术模块与模型性能,进行对比分析。归纳总结轻量级深度学习网络在果实目标检测、谷物穗部检测、作物病虫害检测3大类应用场景的研究进展。指出轻量级深度学习网络在农作物目标检测应用上还存在普适性数据集稀缺、模型泛化能力弱、模型检测精度与检测效率的平衡难以把握等问题,并展望通过农业数据集数量、质量与多样性提升,轻量级深度学习网络结构优化,迁移学习应用以及边缘设备硬件加速技术等进一步提升目标检测性能。With the development of deep learning network model applications in the field of computer vision,the performance of object detection in various agricultural scenarios has been greatly boosted.Unlike large-scale deep learning networks deployed in cloud servers,lightweight deep learning networks,due to their smaller number of parameters and computing power,show potential in agricultural scenarios with limited hardware resources and higher real-time requirements,such as the object detection of fruit and vegetable picking robots,object detection of crop pests and weeds,and crop phenotyping,among other tasks.We provide an overview of the model structure,key technology modules and model performance of the current mainstream lightweight deep learning networks,and conduct a comparative analysis;summarize the research progress of lightweight deep learning networks in three major application scenarios,namely,fruit object detection,grain spike detection,and crop pest and disease detection;and analyze the scarcity of universal datasets,the weakness of the model generalization ability,the accuracy and efficiency of model detection,and the lack of model generalization ability in the application of lightweight deep learning networks in the detection of crop targets.It also analyses the scarcity of universal datasets,weak model generalization ability,and difficulty in balancing model detection accuracy and detection efficiency in crop object detection applications,and looks forward to further improving the object detection performance through the enhancement of agricultural datasets in terms of quantity,quality,and diversity,the optimization of the structure of the lightweight deep learning network,the application of migration learning,and the hardware acceleration technology of edge devices.

关 键 词：农作物 目标检测 深度学习 轻量级网络 边缘计算 

分 类 号：S5[农业科学—作物学] TP391.4[自动化与计算机技术—计算机应用技术]