基于卷积神经网络的高光谱图像分类方法

被引：7

作者：

路易 ^{[1
]}

吴玲达 ^{[2
]}

朱江 ^{[1
]}

机构：

[1] 装备学院研究生管理大队

[2] 装备学院复杂电子系统仿真实验室

来源：

计算机工程与设计 | 2018年 / 39卷 / 09期

关键词：

卷积神经网络; 高光谱图像分类; 虚拟样本; 循环学习率; 动量批处理梯度下降;

D O I：

10.16208/j.issn1000-7024.2018.09.025

中图分类号：

TP751 [图像处理方法]; TP183 [人工神经网络与计算];

学科分类号：

摘要：

为解决在样本有限的情况下高光谱图像分类精度不高的问题,提出一种基于卷积神经网络的高光谱遥感图像分类方法。引入滤波、增加虚拟样本、标准化等预处理技术,使分类模型对地物样本种类和数量的敏感度降低;通过对梯度下降法和学习率计算方法进行优化,降低计算复杂度和计算时间;设计符合高光谱数据特点的网络结构,提高分类方法的泛化性。实验结果表明,与传统分类方法进行比较,该方法有较高的分类精度。

引用

页码：2836 / 2841

页数：6

共 9 条

[1] Unsupervised Deep Feature Extraction for Remote Sensing Image Classification. Romero A,Gatta C,Camps-Valls G. IEEE Transactions on Geoscience&Remote Sensing . 2015
[2] Deep Convolutional Neural Networks for Hyperspectral Image Classification[J] . Wei Hu,Yangyu Huang,Li Wei,Fan Zhang,Hengchao Li,Tianfu Wu. &nbspJournal of Sensors . 2015
[3] Hyperspectral image denoising employing a spectral-spatial adaptive total variation model. Yuan, Qiangqiang,Zhang, Liangpei,Shen, Huanfeng. IEEE Transactions on Geoscience and Remote Sensing . 2012
[4] Deep convolutional networks with superpixel segmentation for hyperspectral image classification. Cao J,Chen Z,Wang B. IEEE Geoscience and Remote Sensing Letters . 2016
[5] Combining support vector machines and Markov random fields in an integrated framework for contextual image classification. MOSER G,SERPICO S B. IEEE Transactions on Geoscience and Remote Sensing . 2013
[6] Hyperspectral Image Classification Using Deep Pixel-pair Features. Li W,Wu G,Zhang F,et al. IEEE Transactions on Geoscience and Remote Sensing . 2017
[7] Modeling high dimensional frequency spectral data based on virtual sample generation technique. J.Tang,M.Y.Jia,Z.Liu. IEEE International Conference on Information and Automation . 2015
[8] A survey of virtual sample generation technology for face recognition. L.Li,Y.Peng,G.Qiu,Z.Sun,S.Liu. Artificial Intelligence . 2017
[9] Deep feature extraction and classification of hyperspectral images based on convolutional neural networks. Chen Y S,Jiang H L,Li C Y,et al. IEEE Transactions on Geoscience and Remote Sensing . 2016

← 1 →