基于机器学习的多源遥感影像融合土地利用分类研究

陈磊士,赵俊三,李易,朱祺夫,许可

doi:10.13718/j.cnki.xsxb.2018.10.018

西南师范大学学报(自然科学版)

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
	姓名不能为空！
邮箱
	邮箱不能为空！非法的邮箱地址。
手机号码
	电话不能为空！请输入有效手机号!
标题
	标题不能为空！
留言内容
	内容不能为空！
验证码
	验证码不能为空！验证码错误！

基于机器学习的多源遥感影像融合土地利用分类研究

陈磊士,赵俊三,李易,朱祺夫,许可. 基于机器学习的多源遥感影像融合土地利用分类研究[J]. 西南师范大学学报(自然科学版), 2018, 43(10): 103-111. doi: 10.13718/j.cnki.xsxb.2018.10.018

引用本文:

CHEN Lei-shi, ZHAO Jun-san, LI Yi, ZHU Qi-fu, XU Ke. On Land Use Classification by Means of Machine Learning Based on Multi-source Remote Sensing Image Fusion[J]. Journal of Southwest China Normal University(Natural Science Edition), 2018, 43(10): 103-111. doi: 10.13718/j.cnki.xsxb.2018.10.018

Citation:

CHEN Lei-shi, ZHAO Jun-san, LI Yi, ZHU Qi-fu, XU Ke. On Land Use Classification by Means of Machine Learning Based on Multi-source Remote Sensing Image Fusion[J]. Journal of Southwest China Normal University(Natural Science Edition), 2018, 43(10): 103-111. doi: 10.13718/j.cnki.xsxb.2018.10.018

基于机器学习的多源遥感影像融合土地利用分类研究

陈磊士,赵俊三,李易,朱祺夫,许可

昆明理工大学国土资源工程学院, 昆明 650093

On Land Use Classification by Means of Machine Learning Based on Multi-source Remote Sensing Image Fusion

CHEN Lei-shi, ZHAO Jun-san, LI Yi, ZHU Qi-fu, XU Ke

Faculty of Land and Resource Engineering, Kunming University of Science and Technology, Kunming 650093, China

摘要: 为了快速获取准确的城市土地利用信息，提高西南地区遥感影像城市土地利用分类信息提取的精度，探讨了当前快速发展的机器学习技术在该领域中的分类实验.选用昆明市主城区作为研究区域，以Landsat8与Sentinel-1A影像为原始数据，使用GS变换法对影像进行融合，使用卷积神经网络（Convolutional Neural Network，CNN）和BP神经网络（Back Propagation Network）2种分类算法对融合前后的遥感影像进行土地利用分类信息提取，对分类结果进行分析.研究结果表明：基于Landsat8和Sentinel-1A的融合影像数据的卷积神经网络分类算法具有最好的分类效果，其总体分类精度和Kappa系数分别为85.8091%，0.8124，认为基于多源遥感影像融合的卷积神经网络分类方法是获取准确的城市土地利用分类信息的一种可行的方法，可以为高原地区城市的土地利用分类提取研究参考.
- 机器学习 /
- 城市土地利用分类 /
- 影像融合 /
- 卷积神经网络 /
- Landsat8 /
- Sentinel-1A
Abstract: In order to obtain accurate urban land use information quickly and improve the precision of land use classification information in high altitude areas by remote sensing image, the study deals with the exploration of the application of a rapidly developing technology, machine learning, in such fields. The main urban area of Kunming City was chosen as the case area in the research, taking Landsat8 and Sentinel-1A remote sensing image as the original data. Then the convolution neural network and BP neural network was used to extract the land use classification information of the remote sensing images before and after the fusion. After that the classification results were analyzed. Finally the results show that the classification method of convolutional neural network classification based on fused image data of Land sat 8 and Sentinel-1A had the best classification results, those overall classification accuracy and the Kappa coefficient reached 85.8091% and 0.8124. Therefore the classification method of convolutional neural network based on multi-source remote sensing image fusion is feasible to obtain accurate urban land use classification information, which provides a reference for the research of land use classification in urban areas of high altitude.
- machine learning /
- urban land use classification /
- image fusion /
- convolution neural network /
- Landsat8 /
- Sentinel-1A .

[1]	TOWNSHEND J,MASEK J, HUANG C Q,et al.Global Characterization and Monitoring of Forest Cover Using Landsat Data:Opportunities and Challenges[J].International Journal of Digital Earth,2012,5(5):373-397.
[2]	杜国明, 匡文慧, 孟凡浩, 等. 巴西土地利用/覆盖变化时空格局及驱动因素[J].地理科学进展,2015, 34(1):73-82.
[3]	彭立, 杨武年, 黄瑾. 川西高原多时相干涉雷达土地覆盖分类研究[J]. 西南大学学报(自然科学版), 2016, 38(5):125-132.
[4]	CHEN Y,SU W,LI J,et al.Hierarchical Object Oriented Classification Using Very High Resolution Imagery and LIDAR Data over Urban Areas[J].Advances in Space Research,2009,43(7):1101-1110.
[5]	杨朝斌, 张树文, 卜坤, 等. 高分辨率遥感影像在城市LUCC中的应用[J].中国科学院大学学报,2016, 33(3):289-297.
[6]	蒋楠, 李卫国, 杜培军. 不同遥感数据融合方法在南方水稻面积监测中的应用研究[J]. 西南大学学报(自然科学版), 2012, 34(6):18-24.
[7]	李章成, 李源洪, 周华茂. 基于ALOS_PALSAR双极化雷达影像遥感监测水稻的研究:以德阳地区为例[J]. 西南师范大学学报(自然科学版), 2012, 37(6):62-67.
[8]	赵有松, 李京伟, 陈军. 基于ETM+制作土地利用覆盖图——以制作北京1:5万土地利用覆盖图为例[J].测绘科学,2001, 26(3):3,39-42.
[9]	翟天林, 金贵, 邓祥征, 等. 基于多源遥感影像融合的武汉市土地利用分类方法研究[J].长江流域资源与环境,2016, 25(10):1594-1602.
[10]	吴健生, 潘况一, 彭建, 等. 基于QUEST决策树的遥感影像土地利用分类——以云南省丽江市为例[J].地理研究,2012, 31(11):1973-1980.
[11]	冯丽英. 基于深度学习技术的高分辨率遥感影像建设用地信息提取研究[D]:杭州:浙江大学,2017.
[12]	秦高峰. 基于机器学习的多光谱遥感影像分类及城市扩展研究[D]:重庆:重庆大学,2012.
[13]	曹兆伟, 林宁, 徐文斌, 等. 基于BP神经网络的东屿岛遥感影像分类[J].海洋通报,2016, 35(5):587-593.
[14]	POWELL R L, ROBERTS D A, DENNISON P E,et al.Sub-pixel Mapping of Urban Land Cover Using Multiple Endmember Spectral Mixture Analysis:Manaus,Brazil[J].Remote Sensing of Environment,2007,106(2):253-267.
[15]	MAAD R, HJERTAKER B T, JOHANSEN G A.Semi-empirical Scatter Correction Model for High-speed Gamma-ray Tomography[J].Measurement Science and Technology,2008,19(9):094016.
[16]	DU Q,FOWLER J E, ZHU W.On the Impact of Atmospheric Correction on Lossy Compression of Multispectral and Hyperspectral Imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2009,47(1):130-132.
[17]	张治清, 何宗. GEOEYE-1多光谱与全色影像融合的适应性及质量评价研究[J]. 西南师范大学学报(自然科学版), 2011, 36(1):203-208.
[18]	常亮, 邓小明, 周明全, 等. 图像理解中的卷积神经网络[J].自动化学报,2016, 42(9):1300-1312.
[19]	KARIMI D,AKBARIZADEHG, KANGZAN K,et al.Effective Supervised Multiple-feature Learning for Fused Radar and Optical Data Classification[J].Iet Radar Sonar and Navigation,2017,11(5):768-777.
[20]	王巍, 郑新奇, 原智远, 等. 邻域规则下的遥感图像分类后处理方法研究[J].测绘通报,2015(S2):17-21.

计量

文章访问数: 784
HTML全文浏览数: 351
PDF下载数: 34
施引文献: 0

基于机器学习的多源遥感影像融合土地利用分类研究

陈磊士,赵俊三,李易,朱祺夫,许可

昆明理工大学国土资源工程学院, 昆明 650093

收稿日期: 2018-02-06

关键词:

摘要: 为了快速获取准确的城市土地利用信息，提高西南地区遥感影像城市土地利用分类信息提取的精度，探讨了当前快速发展的机器学习技术在该领域中的分类实验.选用昆明市主城区作为研究区域，以Landsat8与Sentinel-1A影像为原始数据，使用GS变换法对影像进行融合，使用卷积神经网络（Convolutional Neural Network，CNN）和BP神经网络（Back Propagation Network）2种分类算法对融合前后的遥感影像进行土地利用分类信息提取，对分类结果进行分析.研究结果表明：基于Landsat8和Sentinel-1A的融合影像数据的卷积神经网络分类算法具有最好的分类效果，其总体分类精度和Kappa系数分别为85.8091%，0.8124，认为基于多源遥感影像融合的卷积神经网络分类方法是获取准确的城市土地利用分类信息的一种可行的方法，可以为高原地区城市的土地利用分类提取研究参考.

English Abstract

On Land Use Classification by Means of Machine Learning Based on Multi-source Remote Sensing Image Fusion

昆明理工大学国土资源工程学院, 昆明 650093

Received Date: 2018-02-06

Keywords:

Abstract: In order to obtain accurate urban land use information quickly and improve the precision of land use classification information in high altitude areas by remote sensing image, the study deals with the exploration of the application of a rapidly developing technology, machine learning, in such fields. The main urban area of Kunming City was chosen as the case area in the research, taking Landsat8 and Sentinel-1A remote sensing image as the original data. Then the convolution neural network and BP neural network was used to extract the land use classification information of the remote sensing images before and after the fusion. After that the classification results were analyzed. Finally the results show that the classification method of convolutional neural network classification based on fused image data of Land sat 8 and Sentinel-1A had the best classification results, those overall classification accuracy and the Kappa coefficient reached 85.8091% and 0.8124. Therefore the classification method of convolutional neural network based on multi-source remote sensing image fusion is feasible to obtain accurate urban land use classification information, which provides a reference for the research of land use classification in urban areas of high altitude.