Please wait a minute...
浙江大学学报(理学版)  2023, Vol. 50 Issue (1): 83-95    DOI: 10.3785/j.issn.1008-9497.2023.01.012
地球科学     
融合遥感与社会感知数据的城市土地利用分类方法
吴郁文1,2,林杰1,2()
1.浙江大学 地球科学学院,浙江 杭州 310027
2.浙江大学 地理与空间信息研究所,浙江 杭州 310027
Integrating remotely sensed and social sensed data for urban land use classification
Yuwen WU1,2,Jie LIN1,2()
1.School of Earth Science,Zhejiang University,Hangzhou 310027,China
2.Institute of Geography and Spatial Information,Hangzhou 310027,China
 全文: PDF(10002 KB)   HTML( 15 )
摘要:

传统的土地利用分类方法大多基于对资料或影像的人工解译,存在一定的局限性。近年来,结合空间大数据和自然语言处理技术进行低成本快速的土地资源管理已成为研究热点。以美国纽约市曼哈顿区为例,提出了融合遥感影像和社会感知数据的城市土地利用分类方法。从遥感影像中提取光谱特征、从推特数据中提取用户活动时空和主题特征,基于随机森林法和深度神经网络法,构建了细粒度的城市土地利用分类模型。通过对比不同特征组合分类方法的精度,得到结合光谱特征和用户活动时空、主题特征的深度神经网络方法的结果最优,总体精度达82.65%,Kappa系数为70.1%。结果表明,社会感知数据中隐含的用户活动时空模式和活动主题信息均有助于提高城市土地利用分类的精度,而神经网络法可有效融合多源数据,为快速、低成本获取城市土地利用信息提供了新的途径。

关键词: 土地利用分类遥感社会感知随机森林深度神经网络    
Abstract:

Traditional land use classification methods are mostly based on labor-intensive interpretation of image, which have certain limitations. In recent years, integrating big data and natural language processing technology to carry out low-cost and rapid land resource management has become a hot issue. Take Manhattan as an example, this paper studies the urban land use classification based on remotely sensed and social sensed data. The spectral features of remotely sensed image, the spatiotemporal pattern of twitter user trajectory and the latent topics of tweet content related to user activity are extracted. Two common classification methods, random forest and deep neural network, are applied to construct urban land use classification models. The highest accuracy is obtained by deep neural network method based on remotely sensed and social sensed data, with overall accuracy at 82.65%, and Kappa at 70.1%. The results show that both spatiotemporal and textual features extracted from social sensed data are of great importance in urban land use classification. And deep neural network can integrate information from multi-source data, which provides a potential way for effectively classifying urban land use with open-source data.

Key words: land use classification    remotely sensed    social sensed    random forest    deep neural network
收稿日期: 2021-12-08 出版日期: 2023-01-17
CLC:  P 237  
基金资助: 国家自然科学基金资助项目(41501423)
通讯作者: 林杰     E-mail: jielin@zju.edu.cn
作者简介: 吴郁文(1996—),ORCID:https://orcid.org/0000-0002-8726-6287,女,硕士,主要从事时空地理数据分析与建模研究.
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  
吴郁文
林杰

引用本文:

吴郁文, 林杰. 融合遥感与社会感知数据的城市土地利用分类方法[J]. 浙江大学学报(理学版), 2023, 50(1): 83-95.

Yuwen WU, Jie LIN. Integrating remotely sensed and social sensed data for urban land use classification. Journal of Zhejiang University (Science Edition), 2023, 50(1): 83-95.

链接本文:

https://www.zjujournals.com/sci/CN/10.3785/j.issn.1008-9497.2023.01.012        https://www.zjujournals.com/sci/CN/Y2023/V50/I1/83

图1  城市土地利用分类框架
指标名称计算式特征描述
用户总数Up地块到访用户总数
用户日均停留时长Sp=u=1Upd=1Dp,uTp,u,dDp,uUp地块到访用户平均日停留时长
夜间停留比例PoNS=pTn,pTp地块在18:00至次日6:00被访问的停留时长比例
周末停留比例PoESp=Te,pTp地块在周末被访问的停留时长比例
发推熵TEp=-TWp.uTWplnTWp,uTWp地块到访用户访问该地块的次数占该用户所有访问次数比例的熵的总和
访问熵DVEp=-Dp,uDplnDp,uDp地块到访用户访问该地块的天数占该地块被访问天数的比例的熵的总和
表1  用户活动时空特征指标
图2  BP神经网络模型结构
图3  不同特征组合和分类方法的分类精度比较用户精度, 生产者精度。
真实类别预测类别
政府文化商务商业教育工业医疗公园住宅交通空地
政府机关用地(公管)10452100022210
文化设施用地(文化)0021100002500
商务办公用地(商务)00105168000023400
商业服务用地(商业)11751 04130031 47030
教育用地(教育)0062750019300
工业仓储用地(工业)00625037002200
医疗卫生用地(医疗)003900204100
公园与绿地(公园)0012000442630
住宅用地(住宅)002949840044 01930
交通运输用地(交通)009303000128200
待建成地(空地)000101003111
表2  仅用光谱特征作为输入向量的分类结果混淆矩阵
真实类别预测类别
政府文化商务商业教育工业医疗公园住宅交通空地
McNemar's卡方检验22.1*24.5*108*452*10.5*13.2*5*31.04685*0.931.92
政府机关用地(公管)250495000115510
文化设施用地(文化)01221200111000
商务办公用地(商务)003975010005900
商业服务用地(商业)30601 940100358640
教育用地(教育)10326140148300
工业仓储用地(工业)10115044002900
医疗卫生用地(医疗)1001210303800
公园与绿地(公园)000900055840
住宅用地(住宅)902856830023 94241
交通运输用地(交通)508422001109221
待建成地(空地)000200001113
表3  用光谱特征和用户活动时空特征作为输入向量的分类结果混淆矩阵
图4  典型用户活动时空特征在各土地利用类型上的分布
真实类别预测类别
政府文化商务商业教育工业医疗公园住宅交通空地
McNemar's卡方检验50.8*33.2*29.8*142*38.4*36*19.6*1.9842.9*39.3*3
政府机关用地(政府)1122148100011610
文化设施用地(文化)030020000600
商务办公用地(商务)103477510008300
商业服务用地(商业)0091 986000060200
教育用地(教育)0005720105400
工业仓储用地(工业)00112046003010
医疗卫生用地(医疗)0003102602500
公园与绿地(公园)000400062820
住宅用地(住宅)40326010014 28710
交通运输用地(交通)10131000168871
待建成地(空地)00040003703
表4  用光谱和用户活动时空、主题特征作为输入向量的分类结果混淆矩阵
图5  不同特征组合的分类结果局部分析
图6  特征重要度比较
图7  在土地利用分类中特征的重要度比较
1 王协, 章孝灿, 苏程. 基于多尺度学习与深度卷积神经网络的遥感图像土地利用分类[J]. 浙江大学学报(理学版), 2020, 47(6): 715-723. DOI:10.3785/j.issn.1008-9497.2020.06.009
WANG X, ZHANG X C, SU C, et al. Land use classification of remote sensing images based on multi-scale learning and deep convolution neural network[J]. Journal of Zhejiang University (Science Edition), 2020, 47(6): 715-723. DOI:10.3785/j.issn.1008-9497.2020.06.009
doi: 10.3785/j.issn.1008-9497.2020.06.009
2 周珂, 杨永清, 张俨娜, 等. 光学遥感影像土地利用分类方法综述[J]. 科学技术与工程, 2021, 21(32): 13603-13613. DOI:10.3969/j.issn.1671-1815.2021. 32.001
ZHOU K, YANG Y Q, ZHANG Y N, et al. Review of land use classification methods based on optical remote sensing images[J]. Science Technology and Engineering, 2021, 21(32): 13603-13613. DOI:10. 3969/j.issn.1671-1815.2021.32.001
doi: 10. 3969/j.issn.1671-1815.2021.32.001
3 JOZDANI S E, JOHNSON B A, CHEN D. Comparing deep neural networks, ensemble classifiers, and support vector machine algorithms for object-based urban land use/land cover classification[J]. Remote Sensing, 2019, 11(14): 1713. DOI:10.3390/rs11141713
doi: 10.3390/rs11141713
4 LI X T, HU T Y, GONG P, et al. Mapping essential urban land use categories in Beijing with a fast area of interest (AOI)-based method[J]. Remote Sensing, 2021, 13(3): 477. DOI:10.3390/rs13030477
doi: 10.3390/rs13030477
5 LIU Y, LIU X, GAO S, et al. Social sensing: A new approach to understanding our socioeconomic environments[J]. Annals of the Association of American Geographers, 2015, 105(3): 512-530. DOI:10.1080/00045608.2015.1018773
doi: 10.1080/00045608.2015.1018773
6 陈子龙, 王芳, 李少英, 等. 基于多源数据的县域主导功能类型划分及其空间结构模式识别[J]. 地球信息科学学报, 2021, 23(12): 2215-2231. doi:10.12082/dqxxkx.2021.210050
CHEN Z L, WANG F, LI S Y, et al. Classification of county leading function types and pattern recognition of its spatial structure based on multi-source data[J]. Journal of Geo-Information Science, 2021, 23(12): 2215-2231. doi:10.12082/dqxxkx.2021.210050
doi: 10.12082/dqxxkx.2021.210050
7 JIANG Y Q, HUANG X, LI Z L. Spatiotemporal patterns of human mobility and its association with land use types during COVID-19 in New York city[J]. ISPRS International Journal of Geo-Information, 2021, 10(5): 344. DOI:10.3390/ijgi10050344
doi: 10.3390/ijgi10050344
8 KOZLOWSKA A, STEINNOCHER K. Urban activity detection using geo-located Twitter data[J]. GI_Forum, 2020, 2020(8): 15-31. doi:10.1553/giscience2020_01_s15
doi: 10.1553/giscience2020_01_s15
9 IRANMANESH A, CÖMERT N Z, HOŞKARA Ş Ö. Reading urban land use through spatio-temporal and content analysis of geotagged Twitter data[J]. GeoJournal, 2021: 1-18. DOI:10.1553/giscience2020_01_s15
doi: 10.1553/giscience2020_01_s15
10 王润泽, 周鹏, 潘悦, 等. 基于大数据的城市功能区人口时空聚散模式研究[J]. 地理与地理信息科学, 2022, 38(1): 45-50. DOI:10.3969/j.issn.1672-0504. 2022.01.007
WANG R Z, ZHOU P, PAN Y, et al. Study on spatiotemporal aggregation and dispersion patterns of population in different urban functional areas based on big data[J]. Geography and Geo-Information Science, 2022, 38(1): 45-50. DOI:10.3969/j.issn. 1672-0504.2022.01.007
doi: 10.3969/j.issn. 1672-0504.2022.01.007
11 YIN J J, CHI G Q. Characterizing people's daily activity patterns in the urban environment: A mobility network approach with geographic context-aware twitter data[J]. Annals of the American Association of Geographers, 2021, 111(7): 1967-1987. DOI:10.1080/24694452.2020.1867498
doi: 10.1080/24694452.2020.1867498
12 CHEN B, XU B, GONG P. Mapping essential urban land use categories (EULUC) using geospatial big data: Progress, challenges, and opportunities[J]. Big Earth Data, 2021, 5(3): 410-441. DOI:10.1080/24694452.2020.1867498
doi: 10.1080/24694452.2020.1867498
13 ZHAI W, BAI X Y, SHI Y, et al. Beyond word2vec: An approach for urban functional region extraction and identification by combining place2vec and POIs[J]. Computers, Environment and Urban Systems, 2019, 74: 1-12. DOI:10.1016/j.compenvurbsys.2018.11.008
doi: 10.1016/j.compenvurbsys.2018.11.008
14 ANDRADE R, ALVES A, BENTO C. POI mining for land use classification: A case study[J]. ISPRS International Journal of Geo-Information, 2020, 9(9): 493. doi:10.3390/ijgi9090493
doi: 10.3390/ijgi9090493
15 吴琳琳, 李晓燕, 毛德华, 等. 基于遥感和多源地理数据的城市土地利用分类[J]. 自然资源遥感, 2022, 34(1): 127-134. DOI:10.6046/zrzyyg.2021061
WU L L, LI X Y, MAO D H, et al. Urban land use classification based on remote sensing and multi-source geographic data[J]. Remote Sensing for Natural Resources, 2022, 34(1): 127-134. DOI:10. 6046/zrzyyg.2021061
doi: 10. 6046/zrzyyg.2021061
16 TIAN H C, ZHANG M, LUO X Y, et al. Twitter user location inference based on representation learning and label propagation[C]// Proceedings of the Web Conference 2020. New York: Association for Computing Machinery, 2020: 2648-2654. DOI:10. 1145/3366423.3380019
doi: 10. 1145/3366423.3380019
17 HÄBERLE M, WERNER M, ZHU X X. Geo-spatial text-mining from Twitter: A feature space analysis with a view toward building classification in urban regions[J]. European Journal of Remote Sensing, 2019, 52(supp2): 2-11. DOI:10.1080/22797254.2019.1586451
doi: 10.1080/22797254.2019.1586451
18 FALCONE D, MASCOLO C, COMITO C, et al. What is this place? Inferring place categories through user patterns identification in geo-tagged tweets[C]// 6th International Conference on Mobile Computing, Applications and Services. Austin: IEEE, 2014: 10-19. DOI:10.4108/icst.mobicase. 2014.257683
doi: 10.4108/icst.mobicase. 2014.257683
19 CUI R H, AGRAWAL G, RAMNATH R. Tweets can tell: Activity recognition using hybrid long short-term memory model[C]// Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining. Vancouver: Association for Computing Machinery, 2019: 164-167. doi:10.1145/3341161.3342935
doi: 10.1145/3341161.3342935
20 LEE K, GANTI R K, SRIVATSA M, et al. When twitter meets foursquare: Tweet location prediction using foursquare[C]// Proceedings of the 11th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services. London: ICST, 2014: 198-207. doi:10.4108/icst.mobiquitous.2014.258092
doi: 10.4108/icst.mobiquitous.2014.258092
21 HALIMI A, AYDAY E. Profile matching across online social networks[C]// International Conference on Information and Communications Security. Copenhagen: Springer, 2020: 54-70. doi:10.1007/978-3-030-61078-4_4
doi: 10.1007/978-3-030-61078-4_4
22 RAMAGE D, HALL D, NALLAPATI R, et al. Labeled LDA: A supervised topic model for credit attribution in multi-labeled corpora[C]// Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing. Singapore: Association for Computational linguistics, 2009: 248-256. DOI:10.5555/1699510.1699543
doi: 10.5555/1699510.1699543
23 王瑞, 龙华, 邵玉斌, 等. 基于Labeled-LDA模型的文本特征提取方法[J]. 电子测量技术, 2020, 43(1): 141-146. DOI:10.19651/j.cnki.emt.1903246
WANG R, LONG H, SHAO Y B, et al. Text feature extract method based on Labeled-LDA mode[J]. Electronic Measurement Technology, 2020, 43(1):141-146. DOI:10.19651/j.cnki.emt.1903246
doi: 10.19651/j.cnki.emt.1903246
24 QUERCIA D, ASKHAM H, CROWCROFT J. TweetLDA: Supervised topic classification and link prediction in twitter[C]// Proceedings of the 4th Annual ACM Web Science Conference. 2012: 247-250. DOI:10.1145/2380718.2380750
doi: 10.1145/2380718.2380750
25 朱晓霞, 宁晓刚, 王浩, 等. 高精度地表覆盖数据优化分割的土地利用分类[J]. 测绘科学, 2021, 46(6): 140-149.
ZHU X X, NING X G, WANG G, et al. Land use classification for optimization segmentation based on high-precision land cover data[J]. Science of Surveying and Mapping, 2021, 46(6): 140-149.
26 李敏, 刘国栋, 谭凌. 基于随机森林的土地利用分类与景观格局分析[J]. 地理空间信息, 2022, 20(2): 51-56. DOI:10.3969/j.issn.1672-4623.2022.02.010
LI M, LIU G D, TAN L. Land use classification and landscape pattern analysis based on random forest method[J]. Geospatial Information, 2022, 20(2): 51-56. DOI:10.3969/j.issn.1672-4623. 2022.02.010
doi: 10.3969/j.issn.1672-4623. 2022.02.010
27 段宇英, 汤军, 刘远刚, 等. 基于随机森林的山西省柳林县黄土滑坡空间敏感性评价[J]. 地理科学, 2022, 42(2): 343-351.
DUAN Y Y, TANG J, LIU Y G, et al. Spatial sensitivity evaluation of loess landslide in Liulin county, Shanxi based on random forest[J]. Scientia Geographica Sinica, 2022, 42(2): 343-351.
28 靖娟利, 刘兵, 徐勇, 等. 基于多特征融合的反向传播神经网络高分影像分类与变化检测[J]. 科学技术与工程, 2021, 21(36): 15378-15385. doi:10.3969/j.issn.1671-1815.2021.36.011
JING J L, LIU B, XU Y, et al. High-resolution remote sensing image classification and change detection based on back propagation neural network with multi-feature fusion[J]. Science Technology and Engineering, 2021, 21(36): 15378-15385. doi:10.3969/j.issn.1671-1815.2021.36.011
doi: 10.3969/j.issn.1671-1815.2021.36.011
29 张贝娜, 冯震华, 张丰, 等. 基于时空多视图BP神经网络的城市空气质量数据补全方法研究[J]. 浙江大学学报(理学版), 2019, 46(6): 737-744. DOI:10. 3785/j.issn.1008-9497.2019.06.016
ZHANG B N, FENG Z H, ZHANG F, et al. Urban air quality data completion method based on spatio-temporal multi-view BP neural network[J]. Journal of Zhejiang University (Science Edition), 2019, 46(6): 737-744. DOI:10.3785/j.issn.1008-9497. 2019.06.016
doi: 10.3785/j.issn.1008-9497. 2019.06.016
30 SANDRI M, ZUCCOLOTTO P. A bias correction algorithm for the Gini variable importance measure in classification trees[J]. Journal of Computational and Graphical Statistics, 2008, 17(3): 611-628. DOI:10.1198/106186008X344522
doi: 10.1198/106186008X344522
[1] 朱悦,徐志红,蒋晓敏,张嫣然,王建锋,王善华,陈昱蓉,张丰. 基于遥感生态指数的土地利用程度与环境质量协调发展研究[J]. 浙江大学学报(理学版), 2023, 50(3): 322-331.
[2] 胡东滨,冯婧瑜,杨艺,易国栋. 考虑处置效果的苯系物泄露应急方案生成方法[J]. 浙江大学学报(理学版), 2022, 49(4): 457-466.
[3] 周国华, 蒋晖, 顾晓清, 殷新春. 基于半监督子空间迁移的稀疏表示遥感图像场景分类方法[J]. 浙江大学学报(理学版), 2021, 48(6): 684-693.
[4] 王协, 章孝灿, 苏程. 基于多尺度学习与深度卷积神经网络的遥感图像土地利用分类[J]. 浙江大学学报(理学版), 2020, 47(6): 715-723.
[5] 卢家品, 罗月童, 黄兆嵩, 张延孔, 陈为. 基于排名学习和多源信息的地图匹配方法[J]. 浙江大学学报(理学版), 2020, 47(1): 27-35.
[6] 姜斌, 黄祥志, 杜震洪, 张丰, 刘仁义. 一种动态实时的遥感专题应用系统定制框架[J]. 浙江大学学报(理学版), 2018, 45(6): 758-764.
[7] 吴小君, 方秀琴, 苗月鲜, 王凯, 庞婵云. 基于Landsat的合肥市城市扩张研究[J]. 浙江大学学报(理学版), 2017, 44(6): 631-639.
[8] 江彬彬, 李辉林, 滕国超, 闫金宝, 李宝喜, 李巨宝, 张霄宇, 张宝华. 基于GOCI提取东海近几年赤潮信息及时序分析[J]. 浙江大学学报(理学版), 2017, 44(5): 576-583.
[9] 陈可欣, 张丰, 杜震洪, 刘仁义. 基于遥感影像的嘉兴市城市扩张与驱动力分析[J]. 浙江大学学报(理学版), 2016, 43(6): 709-715.
[10] . 基于GPU加速的遥感影像金字塔创建算法及其在土地遥感影像管理中的应用[J]. 浙江大学学报(理学版), 2011, 38(6): 695-700.