基于深度学习的林业害虫智能化检测方法研究进展  

Research Progress in Deep-Learning-based Intelligent Forest Pest Detection Methods

作　　者：崔晓辰 雷一东[1] Cui Xiaochen;Lei Yidong(Department of Environmental Science and Engineering,Fudan University,Shanghai 200438,China)

机构地区：[1]复旦大学环境科学与工程系,上海200438

出　　处：《世界林业研究》2024年第4期53-57,共5页World Forestry Research

摘　　要：林业是我国重要的基础产业,能将生态效益、经济效益、社会效益集于一体。害虫暴发会对林业资源及生态安全造成极大威胁,给国家造成巨大经济损失。快速、准确、有效地检测林业害虫能够阻止病虫害的扩散蔓延,促进森林病虫害的综合治理,减轻对林业生产和生态环境建设的危害。深度学习方法的飞速发展促进了林业害虫检测精度的提高。文中概述了深度学习的发展进程,通过总结国内外关于深度学习方法在林业害虫智能化检测中的应用,探讨深度学习方法在实际应用过程中存在的问题,并展望其在林业害虫智能化检测方面的发展趋势,以期为林业害虫智能化检测发展提供参考。Forestry is an important industry in China,which integrates ecological,economic and social benefits.Pests pose a great threat to forest resources and ecological security,and bring in huge economic losses to countries.Rapid,accurate and effective detection of forest pests can curb the spread of and promote the integrated control of pests and diseases,and reduce the harm to forests and ecological environment.The rapid development of deep learning methods has improved the accuracy of forest pest detection.This paper outlines the development process of deep learning,summarizes its application in forest pest detection at home and abroad,discusses the problems and their solutions in the application of deep learning methods,and prospects the development trend of deep learning methods in forest pest detection,in order to provide references for the intelligent forest pest detection.

关 键 词：深度学习 目标检测 林业 害虫检测 

分 类 号：S763[农业科学—森林保护学] TP391[农业科学—林学]