深度学习在森林火灾监测领域的应用研究进展——基于CiteSpace的知识图谱分析  

Research Progress in Deep Learning Applications to Forest Fire Monitoring:A Knowledge Graph Analysis Based on CiteSpace

作　　者：刘家豪 任洪 包明伟 刘建祥[1] LIU Jiahao;REN Hong;BAO Mingwei;LIU Jianxiang(College of Soil and Water Conservation,Southwest Forestry University,Kunming 650224,China)

机构地区：[1]西南林业大学水土保持学院,云南昆明650224

出　　处：《玉溪师范学院学报》2024年第6期58-66,共9页Journal of Yuxi Normal University

基　　金：玉溪市新平县第一次森林火灾风险普查项目(YNQY-ZB-2022-90).

摘　　要：深度学习在森林火灾监测中的应用呈现出显著的上升趋势,研究热点从遥感与神经网络逐步转向机器学习和森林管理,体现了火灾监测与评估在多场景应用中的扩展.研究前沿包括边缘识别、特征抽取、多模态框架和生成对抗网络等,这些成果显著提高了火灾监测的精准性和鲁棒性.中国、美国和印度是该领域的主要研究国家,其中中国的研究发表量居首,但国际合作相对较少.从文献研究的角度而言,“深度学习在森林火灾监测领域的研究展望”采用文献计量与分析的方法具有可行性.The analysis reveals a significant upward trend in the application of deep learning for forest fire monitoring,with research hotspots gradually shifting from remote sensing and neural networks to machine learning and forest management,reflecting an expansion of fire monitoring and assessment in multi-scenario applications.Key frontiers of research include edge detection,feature extraction,multimodal frameworks,and generative adversarial networks,which markedly enhance the precision and robustness of fire monitoring.China,the United States,and India are identified as the leading countries in this domain,with China having the highest publication output but relatively limited international collaboration.From the perspective of literature research,the use of bibliometric and analytical methods for“research prospects of deep learning in forest fire monitoring”is feasible.

关 键 词：深度学习 森林火灾 监测 CITESPACE 文献计量分析 

分 类 号：S762[农业科学—森林保护学]