Please wait a minute...
img

官方微信

遥感技术与应用  2020, Vol. 35 Issue (4): 749-758    DOI: 10.11873/j.issn.1004-0323.2020.4.0749
甘肃遥感学会专栏     
基于CNN的不同空间分辨率影像土地覆被分类研究
李宏达(),高小红(),汤敏
青海师范大学地理科学学院,青海省自然地理与环境过程重点实验室,高原科学与可持续发展 研究院,青藏高原地表过程与生态保育教育部重点实验室,青海 西宁 810008
Land Cover Classification for Different Spatial Resolution Images from CNN
Hongda Li(),Xiaohong Gao(),Min Tang
School of Geographical Sciences, Qinghai Normal University, Qinghai Province Key Laboratory of Physical Geography and Environmental Process, Academy of Plateau Science and Sustainability, MOE Key Laboratory of Tibetan Plateau Land Surface Processes and Ecological Conservationm, Xi'ning 810008, China
 全文: PDF(4662 KB)   HTML
摘要:

基于卷积神经网络(Convolutional Neural Networks, CNN)和5种不同空间分辨率的遥感影像,对西宁市东部一区域开展土地覆被分类研究,旨在探索CNN在不同空间分辨率下进行影像分类的差异性和对不同地物的提取能力。为提高样本的选择效率,引入了窗口滑动方法进行辅助选样。研究表明5种不同空间分辨率影像的总体分类精度均达89%以上,Kappa系数达0.86以上,分类精度较高。在所涉及的分辨率尺度范围内,空间分辨率越高,CNN分类结果越精细,并能保持较高的分类精度,表明CNN更适合高空间分辨率影像分类;但同时影像空间分辨率越高,地物表现出较高的类内变异性和低类间差异性,分类精度有降低的趋势。相比较而言,SPOT 6影像的分类精度最高,同时窗口滑动是一种有效的样本辅助选择方法。研究对今后同类工作具有一定的借鉴意义。

关键词: CNNLandsat-8/Sentinel-2A/SPOT-6/GF-2影像土地覆被分类    
Abstract:

Based on convolutional neural networks and five different spatial resolution remote sensing images, the land use/land cover classification study was carried out on a small area in the eastern part of Xining City, aiming at exploring the differences of image classification by CNN with different spatial resolutions and CNN’s ability to extract different features. In order to improve the selection efficiency of the samples, a window sliding method was introduced to assist the samples selection. The research shows that the overall classification accuracy of the five different spatial resolution images is above 89%, the Kappa coefficient is above 0.86. The result further shows that within the resolution scale the higher the resolution, the performance of the CNN classification results for the details is better, and can maintain high classification accuracy, indicating that CNN is more suitable for high spatial resolution images; at the same time, the image spatial resolution is too high, the ground objects exhibit high intra-class variability and low inter-class variability, the classification accuracy tends to decrease. In comparison, CNN has the best classification effect on SPOT 6 images in this study, and window sliding is an effective sample-assisted selection method. This research has certain reference significance for similar research in the future.

Key words: Convolutional Neural Network    Landsat-8/Sentinel-2A/SPOT-6/GF-2 images    Land cover classification
收稿日期: 2019-08-27 出版日期: 2020-09-15
ZTFLH:  TP75  
基金资助: 青海省科技厅自然科学基金项目(2016?ZJ?907)
通讯作者: 高小红     E-mail: 2395789679@qq.com;xiaohonggao226@163.com
作者简介: 李宏达(1995-),男,湖北荆门人,硕士研究生,主要从事遥感应用与地理空间数据分析研究。E?mail:2395789679@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
李宏达
高小红
汤敏

引用本文:

李宏达,高小红,汤敏. 基于CNN的不同空间分辨率影像土地覆被分类研究[J]. 遥感技术与应用, 2020, 35(4): 749-758.

Hongda Li,Xiaohong Gao,Min Tang. Land Cover Classification for Different Spatial Resolution Images from CNN. Remote Sensing Technology and Application, 2020, 35(4): 749-758.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2020.4.0749        http://www.rsta.ac.cn/CN/Y2020/V35/I4/749

图1  研究区位置图
卫星平台所属国家空间分辨率/m空间参考使用光谱波段影像获取时间
GF-2中国4WGS-1984R、G、B、NIR2015-07-28
SPOT-6法国62016-08-08
Sentinel-2A欧空局102016-07-27
Landsat-8美国15/302016-08-07
表1  卫星影像及其参数
图2  卷积运算示意图
图3  最大池化示意图
图4  ReLu函数图像
图5  卷积神经网络结构示意图
图6  标准假彩色影像与CNN分类结果
Landsat-8Lanssat-8(融合后)Sentinel-2ASPOT-6GF-2
生产者精度/%用户精度/%生产者精度/%用户精度/%生产者精度/%用户精度/%生产者精度/%用户精度/%生产者精度/%用户精度/%
林地88.4894.1296.3796.5493.1297.7393.5598.8493.7797.95
耕地89.2978.1393.4287.9894.8398.2193.6891.0596.6091.65
草地78.4387.5990.0887.5786.8278.9394.2480.6093.1775.79
河流76.5696.0884.1199.4589.9496.4097.07100.0090.7298.55
水库坑塘85.71100.0085.1995.8391.03100.0092.86100.0088.6999.62
富营养化水体————57.14100.0096.88100.0098.5391.78————
工业仓储用地87.1093.1094.92100.0099.6093.9899.0194.1796.8984.52
城镇建设用地98.8787.6394.2291.4793.5693.6191.4697.8390.9595.62
总体精度/%89.0192.6492.3193.6592.46
Kappa系数0.860.910.900.920.90
表2  分类精度评价
图7  不同空间分辨率影像下地物生产者精度
图8  不同空间分辨率影像CNN分类细节对比
1 Tian Fuyou, Wu Bingfang, Zeng Hongwei, et al. Identifying Soybean Cropped Area with Sentinel-2 Data and Multi-layer Neural Network[J]. Journal of Geoinformation Science, 2019,21(6):918-927.
1 田富有,吴炳方,曾红伟,等.基于多层神经网络与Sentinel-2数据的大豆种植区识别方法[J].地球信息科学学报,2019,21(6):918-927.
2 Thanh H N T, Doan T M, Tomppo E, et al. Land Use/Land Cover Mapping Using Multitemporal Sentinel-2 Imagery and Four Classification Methods—A Case Study from Dak Nong, Vietnam[J]. Remote Sensing, 2020,12(9):1367. doi:10.3390/rs12091367.
doi: 10.3390/rs12091367
3 Gu Xiaotian, Gao Xiaohong, Ma Huijuan, et al. Comparision of Machine Learning Methods for Land Use/Land Cover Classification in the Complicated Terrain Regions[J]. Remote Sensing Technology and Application, 2019,34(1):57-67.
3 谷晓天,高小红,马慧娟,等.复杂地形区土地利用/土地覆被分类机器学习方法比较研究[J].遥感技术与应用,2019,34(1):57-67.
4 Li Yafei, Dong Hongbin. Classification of Remote-sensing Image based on Convolutional Neural Network[J]. CAAI Transactions on Intelligent Systems,2018,13(4): 550–556.
4 李亚飞,董红斌. 基于卷积神经网络的遥感图像分类研究[J]. 智能系统学报, 2018,13(4):550–556.
5 Masoud M, Bahram S, Mohammad R, et al. Very Deep Convolutional Neural Networks for Complex Land Cover Mapping Using Multispectral Remote Sensing Imagery[J]. Remote Sensing, 2018,10(7):1119. doi:10.3390/rs10071119.
doi: 10.3390/rs10071119
6 Meng Xiangrui, Zhang Shuqing, Zang Shuying. Remote Sensing Classification of Wetland Communities based on Convolutional Neural Networks and High Resolution Images: A Case Study of the Honghe Wetland[J]. Scientia Geographica Sinica, 2018,38(11):1914-1923.
6 孟祥锐,张树清,臧淑英.基于卷积神经网络和高分辨率影像的湿地群落遥感分类——以洪河湿地为例[J].地理科学,2018,38(11):1914-1923.
7 Zhang Wei,Zheng Ke,Tang Ping,et al. Land Cover Classification with Features Extracted by Deep Convolutional Neural Network[J]. Journal of Image and Graphics,2017,22(8):1144-1153.
7 张伟,郑柯,唐娉,等.深度卷积神经网络特征提取用于地表覆盖分类初探[J]. 中国图象图形学报,2017,22(8) : 1144-1153.
8 Cui Gang, Wu Jinsheng, Yu Zhen, et al. Comparison Analysis on Wheat Mapping Using Deep Learning Algorithm from Different Satellite Data Source[J]. Remote Sensing Technology and Application,2019,34(4):748- 755.
8 崔刚,吴金胜,于镇,等. 深度学习对不同分辨率影像冬小麦识别的适用性研究[J].遥感技术与应用,2019,34(4):748-755.
9 Maggiori E, Tarabalka Y, Charpiat G, et al. Fully Convolutional Neural Networks for Remote Sensing Image Classification[C]// 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). IEEE, 2016.
10 Yang Rui,Qi Yuan,Su Yang.U-Net Neural Networks and Its Application in High Resolution Satellite Image Classification[J]. Remote Sensing Technology and Application,2019,34(6):1-8.
10 杨瑞,祁元,苏阳. 深度学习 U-Net方法及其在高分辨卫星影像分类中的应用[J].遥感技术与应用,2019,34(6):1-8.
11 Liu Xuemei, Gao Xiaohong, Jia Wei. Spatio Temporal Change of Urban Heat Island Effect in Xining from Landsat Image[J]. Journal of Earth Environment, 2019, 10(2):166-176.
11 刘雪梅,高小红,贾伟.基于遥感影像的近30年西宁市热岛效应时空变化[J].地球环境学报,2019,10(2):166-176.
12 Wang Can, Wu Xinhui, Li Lianqing, et al. Convolutional Neural Network in Prediction of Soil Moisture Content[J]. Spectroscopy and Spectral Analysis, 2018,38(1):36-41.
12 王璨,武新慧,李恋卿,等.卷积神经网络用于近红外光谱预测土壤含水率[J].光谱学与光谱分析,2018, 38(1):36-41.
13 Tian Ying, Chen Zhuoqi, Hui Fengming, et al. ESA Sentinel-2A/B Satellite: Characteristics and Applications[J]. Journal of Beijing Normal University (Natural Science Edition), 2019,55(1):57-65.
13 田颖,陈卓奇,惠凤鸣,等.欧空局哨兵卫星Sentinel-2A/B数据特征及应用前景分析[J].北京师范大学学报(自然科学版),2019,55(1):57-65.
14 Dong Yunya, Zhang Qian. A Survey of Depth Semantic Feature Extraction of High-resolution Remote Sensing Images based on CNN[J]. Remote Sensing Technology and Application, 2019,34(1):1-11.
14 董蕴雅,张倩.基于CNN的高分遥感影像深度语义特征提取研究综述[J].遥感技术与应用,2019,34(1):1-11.
15 Zhou Feiyan, Jin Linpeng, Dong Jun. Review of Convolutional Neural Network[J]. Chinese Journal of Computers, 2017,40(6):1229-1251.
15 周飞燕,金林鹏,董军.卷积神经网络研究综述[J].计算机学报, 2017,40(6):1229-1251.
16 Huang H, Sun G Y, Zhang X M, et al. Combined Multiscale Segmentation Convolutional Neural Network for Rapid Damage Mapping from Postearthquake Very High-resolution Images[J]. Journal of Applied Remote Sensing,2019,13(2): 022007.doi:10.1117/1.JRS.13.022007.
doi: 10.1117/1.JRS.13.022007
17 Alshehhi R, Marpu P R, Woon W L, et al. Simultaneous Extraction of Roads and Buildings in Remote Sensing Imagery with Convolutional Neural Networks[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 130:139-149.
18 Scherer D, Müller A, Behnke S. Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition[C]//International Conference on Artificial Neural Networks. 2010.
19 Xu L, Choy C S, Li Y W. Deep Sparse Rectifier Neural Network for Speech Denoising[C]//2016 IEEE International Workshop on Acoustic Signal Enhancement (IWAENC). IEEE, 2016.
20 Shu Na, Liu Bo, Lin Weiwei, et al. Survey of Distribute Machine Learning Platforms and Algorithms[J]. Computer Science, 2019,46(3):9-18.
20 舒娜,刘波,林伟伟,等.分布式机器学习平台与算法综述[J].计算机科学,2019,46(3):9-18.
21 Zhu Xiufang, Pan Yaozhong, Zhang Jinshui, et al. The Effects of Training Samples on the Wheat Planting Area Measure Accuracy in TM Scale(Ⅰ):The Accuracy Response of Different Classifiers to Training Samples[J]. Journal of Remote Sensing,2007,11(6):826-837.
21 朱秀芳,潘耀忠,张锦水,等.训练样本对TM尺度小麦种植面积测量精度影响研究(Ⅰ)——训练样本与分类方法间分类精度响应关系研究[J].遥感学报,2007,11(6):826-837.
22 Nogueira K, Penatti Otávio A B, Santos J A D. Towards Better Exploiting Convolutional Neural Networks for Remote Sensing Scene Classification[J]. Pattern Recognition, 2017,61:539-556. doi:10.1016/j.patcog.2016.07.001.
doi: 10.1016/j.patcog.2016.07.001
23 McFEETERS S. K. The Use of the Normalized Difference Water Index (NDWI) in the Delineation of Open Water Features[J]. International Journal of Remote Sensing, 1996,17(7):1425-1432.
24 Ming Dongping, Wang Qun, Yang Jianyu. Spatial Scale of Remote Sensing Image and Selection of Optimal Spatial Resolution[J]. Journal of Remote Sensing, 2008,12(4): 529-537.
24 明冬萍,王群,杨建宇.遥感影像空间尺度特性与最佳空间分辨率选择[J].遥感学报,2008,12(4):529-537.
25 Chen J, Deng M, Xiao P F, et al. Optimal Spatial Scale Choosing for High Resolution Imagery based on Texture Features Frequency Analysis[J]. Journal of Remote Sensing, 2011,15(3):492-511.
26 Yang Yanjun, Tian Qingjiu, Zhan Yulin, et al. Effects of Spatial Resolution and Texture Features on Multi-spectral Remote Sensing Classification[J]. Journal of Geo-information Science, 2018,20(1):99-107.
26 杨闫君,田庆久,占玉林,等.空间分辨率与纹理特征对多光谱遥感分类的影响[J].地球信息科学学报,2018,20(1):99-107.
[1] 柴旭荣,李明,周义,王金风,田庆春. 影像的土地覆被快速分类[J]. 遥感技术与应用, 2020, 35(2): 315-325.
[2] 周壮,李盛阳,张康,邵雨阳. 基于CNN和农作物光谱纹理特征进行作物分布制图[J]. 遥感技术与应用, 2019, 34(4): 694-703.
[3] 谷晓天, 高小红, 马慧娟, 史飞飞, 刘雪梅, 曹晓敏. 复杂地形区土地利用/土地覆被分类机器学习方法比较研究[J]. 遥感技术与应用, 2019, 34(1): 57-67.
[4] 仇江啸, 王效科. 基于高分辨率遥感影像的面向对象城市土地覆被分类比较研究[J]. 遥感技术与应用, 2010, 25(5): 653-661.