摘要
3D Octave卷积模型在高空间-高光谱影像分类中的应用,可以提高多树种分类任务的精度,对提高森林管理的精细化水平具有重要意义。设计了一种结合三维Octave卷积与注意力机制的3DOC-SSAM模型,通过3D Octave卷积和空间—光谱注意力机制,提高了模型的运行效率和分类性能。研究结果表明:①3DOC-SSAM模型总体精度达到99.53%,相对于SVM、ELM、2D-CNN、3D-CNN分别提高了13.86%、18.49%、12.90%和5.36%。且平均精度AA达到99.38%,Kappa系数达0.9947。②小样本训练的情况下,总体精度和平均精度仍然能够达到96.9%和95.52%,高于对比的模型。研究结果为多树种分类任务提供了一个高效且高精度的解决方案,在林业遥感中的应用前景广阔,有助于提升森林资源管理的科学性和可持续性。
The application of 3D Octave convolution model in high spatial-hyperspectral image classification can improve the accuracy of multi-tree species classification tasks,which is of great significance to improve the re⁃finement level of forest management.A 3DOC-SSAM model combining 3D Octave convolution and attention mechanism is designed.Through 3D Octave convolution and spatial-spectral attention mechanism,the opera⁃tion efficiency and classification performance of the model are improved.The results show that:(1)The over⁃all accuracy of the 3DOC-SSAM model reaches 99.53%,which is 13.86%,18.49%,12.90%and 5.36%higher than that of SVM,ELM,2D-CNN and 3D-CNN,respectively.The average accuracy AA reached 99.38%,and the Kappa coefficient reached 0.9947.(2)In the case of small sample training,the overall accu⁃racy and average accuracy can still reach 96.9%and 95.52%,which is higher than the comparison model.The research results provide an efficient and high-precision solution for multi-tree classification tasks,and have broad application prospects in forestry remote sensing,which is helpful to improve the scientificity and sustain⁃ability of forest resource management.
作者
汪明明
陈芸芝
董琰
刘磊
王钰岢
WANG Mingming;CHEN Yunzhi;DONG Yan;LIU Lei;WANG Yu Ke(Digital China Research Institute(Fujian),Key Laboratory of Spatial Data Mining and Information Sharing,Ministry of Education,National Local Joint Engineering Research Center for Integrated Application of Satellite Space Information Technology,Fuzhou University,Fuzhou 3501116,China;Information Management Center of Sinopec Shengli Oilfield Branch,Dongying 257000,China)
出处
《遥感技术与应用》
CSCD
北大核心
2024年第4期897-904,共8页
Remote Sensing Technology and Application
基金
中国科学院战略性先导科技专项子课题(XDA23100503)
福建省水利科技项目(MSK202210)
福建省水利科技项目(MSK202214)
中国石化胜利油田分公司研究项目(YKJ2210)。
关键词
高光谱遥感
无人机
三维Octave卷积
树种分类
小样本
Hyperspectral remote sensing
UAV
3D Octave convolution
Tree species classification
Small sample
