江西科学2025，Vol.43Issue(1)：26-35,10.DOI:10.13990/j.issn1001-3679.2025.01.004
基于卷积神经网络的高光谱图像分类
Hyperspectral Image Classification Based on Convolutional Neural Networks
摘要
Abstract
In recent years,deep learning models,particularly convolutional neural networks(CNNs),have attracted the attention of researchers due to their ability to extract high-lev-el abstract features,and have been widely applied in hyperspectral image classification.This study compared several commonly used convolutional neural network models,and analyzed their advantages and disadvantages.Hyperspectral image classification experiments were conducted based on the Pavia University,Salinas,and WHU-Hi-HongHu datasets.The results show that 3D-CNN and ResNet34 outperformed 1D-CNN and 2D-CNN in terms of classification accuracy,particularly when handling hyperspectral data with complex spatial features.Additionally,differences in the response of different models to different datasets were observed.3D-CNN performs best on the UP and more complex HongHu datasets,while ResNet34 demonstrates an advantage on the SA dataset.It is suggested that the appro-priate network model should be selected based on the characteristics of the dataset in practi-cal applications,and further development of more generalizable algorithms is needed to achieve better classification results.关键词
深度学习/高光谱图像分类/卷积神经网络/残差结构Key words
deeplearning/hyperspectralimage classification/convolutional neuralnetworks/residual structure分类
信息技术与安全科学引用本文复制引用
李子轩,官云兰,王楠,周世健..基于卷积神经网络的高光谱图像分类[J].江西科学,2025,43(1):26-35,10.基金项目
国家自然科学基金项目(42064001) （42064001）
自然资源部环鄱阳湖区域矿山环境监测与治理重点实验室项目(MEMI-2023-15). （MEMI-2023-15）
