Detailed Mapping of Urban Land Use Based on Multi-Source Data: A Case Study of Lanzhou
"> Figure 1
<p>Overview of the study area.</p> "> Figure 2
<p>Technique flow chart.</p> "> Figure 3
<p>Parcel generation process combining OpenStreetMap (OSM) and Gaode roads.</p> "> Figure 4
<p>Relationship between the number of features and classification accuracy (Level II category).</p> "> Figure 5
<p>Comparison of the results of this paper with the results of the essential urban land use categories (EULUC) in the main city of Lanzhou. (<b>a</b>) overview mapping results implemented in this study; (<b>b</b>) overview mapping results of EULUC; (<b>c</b>,<b>d</b>) Google images corresponding to the local mapping areas; (<b>e</b>,<b>f</b>) local details of <a href="#remotesensing-12-01987-f005" class="html-fig">Figure 5</a>a; (<b>g</b>,<b>h</b>) local details of <a href="#remotesensing-12-01987-f005" class="html-fig">Figure 5</a>b.</p> "> Figure 6
<p>Overall accuracy (<b>a</b>) and kappa coefficient (<b>b</b>) of RF model based on different feature combinations.</p> "> Figure 7
<p>User accuracy and producer accuracy of different land use classes based on different feature combinations (Level II category).</p> "> Figure A1
<p>Features with importance greater than 0.005. (<b>a</b>) feature importance based on mean impurity decrease; (<b>b</b>) proportion of the number of features of each type. Note: Mean_B3, Mean_B8, Mean_B11, Mean_B12, Mean_B2, NDVI, NDWI represent mean of green, near-infrared, SWIR 1, SWIR 2, blue, normalized vegetation index, normalized water index, respectively; Luojia1_MIN, Luojia1_MAX, Luojia1_SUM represent the maximum, minimum, and sum of night light intensity, respectively; S1AVV_SUM, S1AVV_MEAN and S1AVH_SUM represent the sum and mean VV and the sum of VH respectively; POI1per, POI3per, and POI7per represent the proportion of residential, commercial, and medical of POIs, POI2sum, POI3sum, POI5sum, POI6sum, POI7sum, POI8sum represent the total number of business, commercial, administrative, educational, medical of POIs, respectively; pop_Sat22 represents the population density at 22:00 on Saturday, and so on for other population density features.</p> ">
Abstract
:1. Introduction
2. Study Area and Data Sources
2.1. Study Area
2.2. Data Sources and Pretreatment
2.2.1. Remote Sensing Data
2.2.2. Internet Data
2.2.3. Road Network Data
3. Method
3.1. Parcel Generation
3.2. Feature Extraction
3.3. Feature Reduction
3.4. Model Building
3.4.1. Random Forest Model
3.4.2. Classification Based on Different Feature Combination
3.4.3. Samples for Training and Validation
4. Results
4.1. Feature Selection Result
4.2. Classification Results
4.2.1. Classification Results Based on Multi-Source Features Using RF Model
4.2.2. Comparison with EULUC Results
4.3. Feature Importance Analysis
4.3.1. Comparison of Classification Accuracy of Different Feature Combinations
4.3.2. Importance of Different Features
5. Discussion
6. Conclusions
- (1)
- The random forest classifier conducted on multi-source data achieves good classification results. Specifically, for level I categories, the overall accuracy is as high as 83.75%, the kappa coefficient is 0.77, for level II, the accuracy of each type reaches more than 70%. Compared with the results of EULUC, the results in this study represent more details of urban land use.
- (2)
- The combination of multi-source features of remote sensing and Internet data is conducive to the improvement of urban land use classification effects. POI features and temporal population density features could improve the accuracy of all classes significantly; remote sensing features could improve the classification accuracy of some land categories.
- (3)
- POI and time-series population density features contribute the most to urban land use classification, followed by spectral features, while night lighting features and backscatter features have the fewest contributions. Compared single-moment population information, the multi-time population has more contributions for urban land use. Remote sensing data is indispensable for the improvement of urban land use classification, although its contribution relatively little compared with other features.
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
Appendix A
Urban Land Use Class | S2 | S2+S1 | S2+Luojia | S2+Easygo | S2+POI | S2+Easygo+POI | POI | Easygo | Easygo+POI | ALL |
---|---|---|---|---|---|---|---|---|---|---|
UA/PA | UA/PA | UA/PA | UA/PA | UA/PA | UA/PA | UA/PA | UA/PA | UA/PA | UA/PA | |
Village | 37.84/37.84 | 32.43/33.33 | 48.65/38.3 | 48.65/46.15 | 35.14/37.14 | 62.16/56.1 | 24.32/34.62 | 32.43/37.5 | 51.35/41.3 | 64.86/61.54 |
Residential | 61.29/49.67 | 66.94/50.3 | 62.90/52.00 | 85.48/70.2 | 74.19/69.7 | 83.06/78.03 | 61.29/75.25 | 88.71/67.9 | 78.23/72.39 | 88.71/79.14 |
Business | 38.46/38.46 | 26.92/41.18 | 42.31/45.83 | 38.46/62.5 | 53.85/66.67 | 73.08/82.61 | 65.38/65.38 | 34.62/56.25 | 73.08/82.61 | 69.23/81.82 |
Commercial | 39.02/29.63 | 36.59/30.61 | 41.46/44.74 | 48.78/54.05 | 60.98/64.10 | 73.17/69.77 | 68.29/65.12 | 41.46/48.57 | 65.85/69.23 | 78.05/72.73 |
Industrial | 46.00/35.94 | 44.00/36.67 | 44.00/37.93 | 72.00/62.07 | 74.00/49.33 | 74.00/61.67 | 88.00/36.97 | 68.00/56.67 | 72.00/56.25 | 76.00/76.00 |
Administrative | 6.67/25.00 | 0/0 | 0/0 | 33.33/33.33 | 46.67/46.67 | 66.67/66.67 | 60.00/45.00 | 33.33/29.41 | 66.67/71.43 | 73.33/64.71 |
Educational | 21.74/28.57 | 26.09/27.27 | 21.74/27.78 | 65.22/61.22 | 67.39/70.45 | 82.61/67.86 | 69.57/69.57 | 71.74/64.71 | 82.61/69.09 | 82.61/76.00 |
Medical | 0/0 | 0/0 | 28.57/100.00 | 28.57/66.67 | 42.86/75.00 | 14.29/66.67 | 42.86/66.67 | 28.57/80.00 | 21.43/60.00 | 28.57/66.67 |
Sport and cultural | 16.67/100 | 0/0 | 16.67/50.00 | 33.33/100 | 0/0 | 16.67/100.00 | 0/0 | 16.67/100 | 16.67/50.00 | 33.33/66.67 |
Greenspace | 75.00/84.00 | 75.00/84.00 | 50/.0082.35 | 75/84 | 78.57/75.86 | 78.57/84.62 | 17.86/55.56 | 14.29/20.00 | 25.00/38.89 | 78.57/91.67 |
Undeveloped | 7.69/100 | 23.08/100 | 23.08/75.00 | 15.38/100 | 0/0 | 0/0 | 0/0 | 7.69/100.00 | 0/0 | 46.10/100.00 |
References
- Brockmann, D.; Helbing, D. The Hidden Geometry of Complex, Network-Driven Contagion Phenomena. Science 2013, 342, 1337–1342. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Foley, J.A.; Ramankutty, N.; Brauman, K.A.; Cassidy, E.S.; Gerber, J.S.; Johnston, M.; Mueller, N.D.; O’Connell, C.; Ray, D.K.; West, P.C.; et al. Solutions for a cultivated planet. Nature 2011, 478, 337–342. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- McKinney, M.L. Urbanization, Biodiversity, and Conservation. BioScience 2002, 52, 883–890. [Google Scholar] [CrossRef]
- Kuang, W.; Zhang, S.; Liu, J.; Shao, Q. Methodology for classifying and detecting intra-urban land use change: A case study of Changchun city during the last 100 years. J. Remote Sens. 2010, 14, 351–363. [Google Scholar] [CrossRef]
- Zhao, P. Reading New Code For Classification Of Urban Land Use And Planning Standards of Development Land: Its Application And Prospects. Planners 2012, 28, 10–16. Available online: http://en.cnki.com.cn/Article_en/CJFDTOTAL-GHSI201202004.htm (accessed on 18 April 2020).
- Alabi, T.; Haertel, M.; Chiejile, S. Investigating the Use of High Resolution Multi-spectral Satellite Imagery for Crop Mapping in Nigeria Crop and Landuse Classification using WorldView-3 High Resolution Multispectral Imagery and LANDSAT8 Data. In Proceedings of the 2nd International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM), Rome, Italy, 26–27 April 2016; SciTePress: Setubal, Portugal, 2016; pp. 109–120. [Google Scholar]
- Venkateshwarlu, C.; Rao, K.G.; Prakash, A. Artificial neural networks in the improvement of spatial resolution of thermal infrared data for improved landuse classification. In Proceedings of the 2nd Grss/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, Berlin, Germany, 22–23 May 2003; IEEE: Piscataway, NJ, USA, 2012; pp. 162–166. [Google Scholar]
- Kishore, D.D.; Gopal, R.K.; Prakash, A. Improvement of effective spatial resolution of thermal infrared data for urban landuse classification. In Proceedings of the IEEE/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, Rome, Italy, 8–9 November 2001; IEEE: Piscataway, NJ, USA, 2002; pp. 332–336. [Google Scholar]
- Wu, S.-S.; Qiu, X.; Usery, E.L.; Wang, L. Using Geometrical, Textural, and Contextual Information of Land Parcels for Classification of Detailed Urban Land Use. Ann. Assoc. Am. Geogr. 2009, 99, 76–98. [Google Scholar] [CrossRef]
- Huang, B.; Zhao, B.; Song, Y. Urban land-use mapping using a deep convolutional neural network with high spatial resolution multispectral remote sensing imagery. Remote Sens. Environ. 2018, 214, 73–86. [Google Scholar] [CrossRef]
- Wang, C.; Wang, A.; Wang, J.; Liu, Z. Land Use Information Extraction in the Inner City of Beijing Based on Object-oriented Classification Method. J. Nat. Resour. 2015, 30, 705–714. [Google Scholar]
- Tu, W.; Cao, J.; Yue, Y.; Shaw, S.-L.; Zhou, M.; Wang, Z.; Chang, X.; Xu, Y.; Li, Q. Coupling mobile phone and social media data: A new approach to understanding urban functions and diurnal patterns. Int. J. Geogr. Inform. Sci. 2017, 31, 2331–2358. [Google Scholar] [CrossRef]
- Zhong, C.; Huang, X.; Arisona, S.M.; Schmitt, G.; Batty, M. Inferring building functions from a probabilistic model using public transportation data. Comput. Environ. Urban. Syst. 2014, 48, 124–137. [Google Scholar] [CrossRef]
- Chen, Y.; Liu, X.; Li, X.; Liu, X.; Yao, Y.; Hu, G.; Xu, X.; Pei, F. Delineating urban functional areas with building-level social media data: A dynamic time warping (DTW) distance based k-medoids method. Landsc. Urban. Plan. 2017, 160, 48–60. [Google Scholar] [CrossRef]
- Ty, H.; Yang, J.; Xuecao, L.; Gong, P. Mapping Urban Land Use by Using Landsat Images and Open Social Data. Remote Sens. 2016, 8, 151. [Google Scholar]
- Zhu, L.F.; Zhou, L.; Liu, Y.; Chen, Y.X. A New Method with POI Data Applied in the Determination of Urban Functional Agglomeration Regions. In Proceedings of the 2019 IEEE 4th International Conference on Big Data Analytics (ICBDA), Suzhou, China, 15–18 March 2019; IEEE: Piscataway, NJ, USA, 2019; pp. 79–82. [Google Scholar]
- Peng, Z.; Sun, Z.; Cheng, Q.; Jiao, H.; Chen, W. Urban Land Use Function Recognition Method Using Sequential Mobile Phone Data. Geomat. Inform. Sci.Wuhan Univ. 2018, 43, 1399–1407. [Google Scholar]
- Yuan, J.; Zheng, Y.; Xie, X. Discovering regions of different functions in a city using human mobility and POIs. In Proceedings of the KDD ‘12: The ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Beijing, China, 12–16 August 2012; Association for Computing Machinery: New York, NY, USA, 2012; pp. 186–194. [Google Scholar]
- Toole, J.L.; Ulm, M.; Bauer, D.; Gonzalez, M.C. Inferring land use from mobile phone activity. In Proceedings of the UrbComp ‘12, the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Beijing, China, 12–16 August 2012; Association for Computing Machinery: New York, NY, USA, 2012; pp. 1–8. [Google Scholar]
- Liu, X.; Long, Y. Automated identification and characterization of parcels with OpenStreetMap and points of interest. Environ. Plan. B 2016, 43, 341–360. [Google Scholar] [CrossRef]
- Liu, Y.; Wang, F.; Xiao, Y.; Gao, S. Urban land uses and traffic ‘source-sink areas’: Evidence from GPS-enabled taxi data in Shanghai. Landsc. Urban. Plan. 2012, 106, 73–87. [Google Scholar] [CrossRef]
- Grippa, T.; Georganos, S.; Zarougui, S.; Bognounou, P.; Diboulo, E.; Forget, Y.; Lennert, M.; Vanhuysse, S.; Mboga, N.; Wolff, E. Mapping Urban Land Use at Street Block Level Using OpenStreetMap, Remote Sensing Data, and Spatial Metrics. ISPRS Int. J. Geo-Inform. 2018, 7, 246. [Google Scholar] [CrossRef] [Green Version]
- Zhang, Y.; Li, Q.; Huang, H.; Wu, W.; Du, X.; Wang, H. The Combined Use of Remote Sensing and Social Sensing Data in Fine-Grained Urban Land Use Mapping: A Case Study in Beijing, China. Remote Sens. 2017, 9, 865. [Google Scholar] [CrossRef] [Green Version]
- Gong, P.; Chen, B.; Li, X.; Liu, H.; Wang, J.; Bai, Y.; Chen, J.; Chen, X.; Fang, L.; Feng, S.; et al. Mapping Essential Urban Land Use Categories in China (EULUC-China): Preliminary results for 2018. Sci. Bull. 2020, 65, 182–187. [Google Scholar] [CrossRef] [Green Version]
- Tu, Y.; Chen, B.; Zhang, T.; Xu, B. Regional Mapping of Essential Urban Land Use Categories in China: A Segmentation-Based Approach. Remote Sens. 2020, 12, 1058. [Google Scholar] [CrossRef] [Green Version]
- Main-Knorn, M.; Pflug, B.; Debaecker, V.; Louis, J. Calibration and Validation plan for the L2A processor and products of the Sentinel-2 mission. In Proceedings of the 36th International Symposium on Remote Sensing of Environment, Berlin, Germany, 11–15 May 2015; pp. 1249–1255. [Google Scholar]
- Mayer, B.; Kylling, A. Technical Note: The libRadtran Software Package for Radiative Transfer Calculations - Description and Examples of Use. Atmos. Chem. Phys. 2005, 5, 1855–1877. [Google Scholar] [CrossRef] [Green Version]
- Li, X.; Xu, H.M.; Chen, X.L.; Li, C. Potential of NPP-VIIRS Nighttime Light Imagery for Modeling the Regional Economy of China. Remote Sens. 2013, 5, 3057–3081. [Google Scholar] [CrossRef] [Green Version]
- Bennett, M.M.; Smith, L.C. Advances in using multitemporal night-time lights satellite imagery to detect, estimate, and monitor socioeconomic dynamics. Remote Sens. Environ. 2017, 192, 176–197. [Google Scholar] [CrossRef]
- Jiang, W.; He, G.; Long, T.; Guo, H.; Yin, R.; Leng, W.; Liu, H.; Wang, G. Potentiality of Using Luojia 1-01 Nighttime Light Imagery to Investigate Artificial Light Pollution. Sensors 2018, 18, 2900. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Luo, L.; Liu, B.; Xuechao, L. Data Quality Assessment and Application Analysis for OpenStreetMap Road Network. Jiangxi Sci. 2017, 35, 151–157. Available online: http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jxkx201701030 (accessed on 6 February 2020).
- Wang, D.; Wang, S.; Huang, C. Comparison of Sentinel-2 imagery with Landsat8 imagery for surface water extraction using four common water indexes. Remote Sens. Land Resour. 2019, 31, 157–165. [Google Scholar]
- Congalton, R.G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens Environ. 1991, 37, 35–46. [Google Scholar] [CrossRef]
- Kononenko, I. Estimating attributes: Analysis and extensions of RELIEF. In Proceedings of the European Conference on Machine Learning: ECML-94, Catania, Italy, 6–8 April 1994; Springer: Berlin/Heidelberg, Germany, 1994; pp. 171–182. [Google Scholar]
- Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Kulkarni, V.Y.; Sinha, P.K. Pruning of Random Forest classifiers: A survey and future directions. In Proceedings of the 2012 International Conference on Data Science & Engineering (ICDSE), Cochin, Kerala, India, 18–20 July 2012; IEEE: Piscataway, NJ, USA, 2012; pp. 64–68. [Google Scholar]
- Department of Natural Rescources of Lanzhou. Available online: http://zrzyj.lanzhou.gov.cn/art/2017/9/19/art_10426_554643.html?authkey=frmul3 (accessed on 14 March 2020).
- Su, M.; Guo, R.; Chen, B.; Hong, W.; Wang, J.; Feng, Y.; Xu, B. Sampling Strategy for Detailed Urban Land Use Classification: A Systematic Analysis in Shenzhen. Remote Sens. 2020, 12, 1497. [Google Scholar] [CrossRef]
- Liu, X.P.; He, J.L.; Yao, Y.; Zhang, J.B.; Liang, H.L.; Wang, H.; Hong, Y. Classifying urban land use by integrating remote sensing and social media data. Int. J. Geogr. Inform. Sci. 2017, 31, 1675–1696. [Google Scholar] [CrossRef]
- Manonmani, R.; Suganya, G.M.D. Remote sensing and GIS application in change detection study in urban zone using multi temporal satellite. Int. J. Geomat. Geosci. 2010, 1, 60–65. [Google Scholar]
- He, S.; Wang, X.; Dong, J.; Wei, B.; Duan, H.; Jiao, J.; Xie, Y. Three-Dimensional Urban Expansion Analysis of Valley-Type Cities: A Case Study of Chengguan District, Lanzhou, China. Sustainability 2019, 11, 5663. [Google Scholar] [CrossRef] [Green Version]
Level I | Level II |
---|---|
Residential | Village, Residential |
Commercial | Business, Commercial |
Industrial | Industrial |
Transportation | Transportation |
Public | Administrative, Educational, Medical, Sport and cultural, Greenspace, Undeveloped |
Feature Information | Parameter | Count |
---|---|---|
Spectral | Mean and standard deviation of red, green, blue, near-infrared, and two short-wave infrared bands; mean of NDVI, NDWI and NDBI | 15 |
Texture | Parameters (mean, standard deviation, homogeneity, contrast, heterogeneity, entropy) based on GLCM of red, green, blue, near-infrared, and two short-wave infrared bands | 36 |
Backscatter | Mean, maximum, minimum, standard deviation, range, and the sum of VV and VH | 12 |
Nighttime light | Mean, maximum, minimum, standard deviation, range, and the sum of light intensity | 6 |
POI | Total number of all POIs Total number of each type of POIs The proportion of each type of POIs | 23 |
Time series population density | Population density values at every two hours in a week | 84 |
Experiment | Feature Combination |
---|---|
S2 | Spectral and texture features |
S2 + S1 | Spectral, texture, and backscatter features |
S2 + Luojia | Spectral, texture, and nighttime light features |
S2 + Easygo | Spectral, texture, and time-series population density features |
S2 + POI | Spectral, texture, and POI features |
S2 + Easygo + POI | Spectral, texture, POI, time-series population density features |
POI | POI features |
Easygo | Time-series population density features |
POI + Easygo | POI and time-series population density features |
All | Spectral, texture, nighttime light, backscatter, POI, and time-series population density features |
Actual Results | Classification Results | UA/% | |||
---|---|---|---|---|---|
Residential | Commercial | Industrial | Public | ||
Residential | 150 | 6 | 1 | 4 | 93.17 |
Commercial | 2 | 54 | 4 | 7 | 80.60 |
Industrial | 8 | 2 | 38 | 2 | 76.00 |
Public | 18 | 4 | 7 | 93 | 76.23 |
PA/% | 84.27 | 81.82 | 76.00 | 87.74 | |
OA = 83.75% kappa coefficient = 0.77 |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zong, L.; He, S.; Lian, J.; Bie, Q.; Wang, X.; Dong, J.; Xie, Y. Detailed Mapping of Urban Land Use Based on Multi-Source Data: A Case Study of Lanzhou. Remote Sens. 2020, 12, 1987. https://doi.org/10.3390/rs12121987
Zong L, He S, Lian J, Bie Q, Wang X, Dong J, Xie Y. Detailed Mapping of Urban Land Use Based on Multi-Source Data: A Case Study of Lanzhou. Remote Sensing. 2020; 12(12):1987. https://doi.org/10.3390/rs12121987
Chicago/Turabian StyleZong, Leli, Sijia He, Jiting Lian, Qiang Bie, Xiaoyun Wang, Jingru Dong, and Yaowen Xie. 2020. "Detailed Mapping of Urban Land Use Based on Multi-Source Data: A Case Study of Lanzhou" Remote Sensing 12, no. 12: 1987. https://doi.org/10.3390/rs12121987