تهیه نقشه بلوک های ساختمانی شهری در سامانه گوگل ارث انجین از طریق آموزش طبقه بندی کننده تعمیم پذیر

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی کارشناسی ارشد فتوگرامتری، دانشگاه صنعتی خواجه نصیرالدین طوسی

2 دانشجوی دکتری تخصصی سنجش از دور، دانشگاه صنعتی خواجه نصیرالدین طوسی

3 استاد گروه فتوگرامتری و سنجش ازدور، دانشگاه صنعتی خواجه نصیرالدین طوسی

4 دانشیار گروه فتوگرامتری و سنجش ازدور، دانشگاه صنعتی خواجه نصیرالدین طوسی

5 استادیار گروه ژئودزی و مهندسی نقشه برداری، دانشگاه تفرش

10.22131/sepehr.2021.247873

چکیده

امروزه فناوری سنجش ازدور جایگاهی ویژه در کاربردهای مختلف مدیریت شهری پیدا کرده است. در این بین، نقشه ی ساختارهای شهری نظیر بلوک های ساختمانی، عموماً در مدیریت بحران، طراحی شهری و مطالعات مربوط به توسعه ی شهری مورد استفاده قرار می گیرند. در این مطالعه تولید نقشه بلوک های ساختمانی با استفاده از تصاویر ماهواره ای سنتینل 1 و 2 دنبال شده است. روش پیشنهادی این مقاله متکی بر استفاده از طبقه بندی کننده آموزش یافته تعمیم پذیر می باشد. به نحوی که در ابتدا، طبقه بندی کننده مورد نظر با استفاده از نمونه های آموزشی به دست آمده از یک فرآیند پالایشی سختگیرانه نوین توسط محصولات سنجش ازدوری و مکانی مختلف، در سال 2015، آموزش می یابد. سپس این طبقه بندی کننده به منظور تولید نقشه بلوک های ساختمانی در مقاطع زمانی مشابه سه سال هدف (2018، 2019 و 2020) به کار گرفته می شود. به دلیل تنوع بافت و تراکم بلوک های ساختمانی در کلان شهر تهران، روش پیشنهاد شده در این منطقه مورد ارزیابی قرار گرفته است. همچنین با توجه به وسعت منطقه مطالعاتی، فراهم بودن تصاویر ماهواره ای رایگان بدون نیاز به اخذ و امکان اجرای عملیات  مختلف پردازشی به صورت برخط، از سامانه گوگل ارث انجین در پژوهش حاضر استفاده شده است. سه روش طبقه بندی جنگل تصادفی، کمترین فاصله با معیار فاصله ماهالانابیس و ماشین بردارپشتیبان در این فرآیند مورد بررسی قرار می گیرند. به منظور ارزیابی روش پیشنهادی، از نمونه های مرجع به دست آمده از تفسیر بصری تصاویر با قدرت تفکیک مکانی بالا (گوگل ارث) در هر سه سال هدف استفاده شده است. نتایج به دست آمده عملکرد بهتر روش جنگل تصادفی در هر سه سال هدف با دقت کلی بالای 93 درصد را نسبت به دو روش دیگر نشان می دهند.

کلیدواژه‌ها


عنوان مقاله [English]

Urban building blocks mapping through a generalizable trained classifier in Google Earth Engine platform

نویسندگان [English]

  • Alireza Taheri Dehkordi 1
  • Seyyed Mohammad Milad Shahabi 2
  • Mohammad Javad Valadan Zouj 3
  • Mahmood Reza Sahebi 4
  • Alireza Safdarinejad 5
1 M. Sc student, Department of Photogrammetry and Remote sensing, K. N. Toosi University of Technology, Tehran, Iran
2 Ph. D student, Department of Photogrammetry and Remote sensing, K. N. Toosi University of Technology, Tehran, Iran
3 Professor, Department of Photogrammetry and Remote sensing, K. N. Toosi University of Technology, Tehran, Iran
4 Associate Professor, Department of Photogrammetry and Remote sensing, K. N. Toosi University of Technology, Tehran, Iran
5 Assistant professor, Department of Geodesy and Surveying Engineering, Tafresh University, Tafresh, Iran
چکیده [English]

Extended Abstract
Introduction
Over the past three decades, with the rapid development of spatial-based satellite imagery, remote sensing technology has found a special place in various applications of urban management. Production of status maps of urban structures, the study of energy loss status, identification of thermal islands, monitoring of urban vegetation, and assessment of air pollution are just a few examples of areas related to urban management that remote sensing technology is the basis for indirect measurement of the related quantities. Maps of urban structures such as building blocks are commonly used in crisis management, urban design, and urban development studies.
 
Materials
In this study, the production of urban building block maps using Sentinel 1 and 2 satellite images has been conducted. Normalized Difference Vegetation Index (NDVI) and Normalized Difference Building Index ( NDBI ) for three consecutive months, the slope feature derived from the 30-meter Shuttle Radar Topographic Mission (SRTM)Digital Elevation Model of the study area, along with two Vertical – Vertical (VV) and Vertical - Horizontal ( VH ) polarization in both ascending and descending orbits, form the set of input features.
 
Methods
The proposed method of this paper relies on the use of a generalizable trained classifier. Initially, the classifier is trained in 2015 using training samples obtained from a new rigorous refining process using different remote sensing and spatial products. This rigorous refining process uses a reference urban map of 2015. In the first step, the corresponding areas related to the ways and roads are removed using the OpenStreetMap data layer. Areas suspected of vegetation with NDVI greater than 0.2 are then discarded. Also, due to the high backscattering of buildings in Synthetic Aperture Radar images, areas with a value less than the average backscattering coefficient of the remaining areas are eliminated. Finally, the residual map is refined using the Mahalanabis distance and the Otsu automatic thresholding method. The trained classifier is then used to generate a map of building blocks at similar time intervals for the three target years (2018, 2019, and 2020). Due to the diversity of texture and density of building blocks in the metropolis of Tehran, the proposed method has been evaluated in this area. Due to the concentration of political, welfare, and social facilities, Tehran has experienced more unplanned and irregular expansion and urbanization than other cities in Iran, which has lead to changes in buildings and constructions. Also, due to the availability of free satellite images and various online processing operations, the Google Earth Engine platform has been used in this study. The performance of three different classifiers including Random Forest (RF), Minimum Mahalanabis Distance (MD), and Support Vector Machines (SVM) are examined in this process. In order to evaluate the proposed method, reference samples obtained from visual interpretation of high-resolution satellite images (Google Earth) in all three target years have been used.
 
Results
The performance of the aforementioned classifiers has been investigated using 3 different criteria: overall accuracy, user accuracy, and F-score of building blocks. The RF method with an overall accuracy of over 93% in all three target years has shown the best performance. The SVM method ranks second with an accuracy of about 91% every three years. However, the MD method with an overall accuracy below 85% in all three target years has not performed well.
 
Discussion
The results show better performance of the RF method in all three target years with an overall accuracy of over 93%. It should be noted that the MD classifier with higher user accuracy than other methods, has shown better performance in detecting the class of building blocks. However, the RF method is the best classifier in terms of the user accuracy of the background class. The effect of using two VV and VH polarization and also the slope derived from the SRTM Model in the input feature set on the final accuracy of classification was also investigated. According to the results, the simultaneous use of these three features produces more accurate results in both target classes. However, the results show that the use of VV polarization increases the final classification accuracy compared to VH polarization. The presence of slope feature along with both polarizations has also increased the classification accuracy of each class, especially the background class. However, the exclusion of both VV and VH features from the input feature set has resulted in a more than 10% reduction in overall classification accuracy.
 
Conclusion
Based on calculated overall accuracies which are above 80% in the majority of investigated cases, two different results can be concluded. First, the trained classifier has shown good temporal generalization and has achieved acceptable accuracy in the target years. Second, due to the different collection processes of training and evaluation data, the proposed rigorous refining method for the preparation of training data has shown good performance. The effect of using two VV and VH polarization and also the slope derived from the SRTM  Digital Elevation Model in the input feature set on the final accuracy of classification was also investigated. According to the results, the simultaneous use of these three features produces more accurate results in both target classes. However, the results show that the use of VV polarization increases the final classification accuracy compared to VH polarization. The presence of slope feature along with both polarizations has also increased the classification accuracy of each class, especially the background class. However, the exclusion of both VV and VH features from the input feature set has resulted in a tangible decreasein overall classification accuracy.
 

کلیدواژه‌ها [English]

  • Remote Sensing
  • Building Blocks
  • Generalizable Trained Classifier
  • Google Earth Engine
  • Sentinel Satellite Images
1- Bartholome, Etienne, & Allan S. Belward. (2005). GLC2000: a new approach to global land cover mapping from Earth observation data. International Journal of Remote Sensing.
2- Bontemps, Sophie, & et al. (n.d.). GLOBCOVER 2009-Products description and validation report. 2011: URL: http://ionia1. esrin. esa. int/docs/GLOBCOVER2009_Validation_Report_2 2.
3- Carrasco, Luis, & et al. (2019). Evaluating combinations of temporally aggregated Sentinel-1, Sentinel-2 and Landsat 8 for land cover mapping with Google Earth Engine. Remote Sensing.
4- Chini, Marco, & et al. (2018). Towards a 20 m global building map from sentinel-1 sar data. Remote Sensing.
5- Corbane, C., Syrris, V., Sabo, F., Politis, P., Melchiorri, M., Pesaresi, M., ... & Kemper, T. (2020). Convolutional
neural networks for global human settlements mapping from Sentinel-2 satellite imagery. Neural Computing and Applications, 1-24.
6- Deepthi, R., S. Ravindranath, & K. G. Raj. (2018). EXTRACTION OF URBAN FOOTPRINT OF BENGALURU CITY USING MICROWAVE REMOTE SENSING. International Archives of the Photogrammetry, Remote Sensing & Spatial Information Sciences.
7- Filipponi, & Federico. (2019). Sentinel-1 GRD Preprocessing Workflow. Multidisciplinary Digital Publishing Institute Proceedings.
8- Gamba, Paolo, & Martin Herold. (2009). Global mapping of human settlement: experiences, datasets, and prospects. CRC Press.
9- Gaughan, Andrea E, & et al. (2013). High resolution population distribution maps for Southeast Asia in 2010 and 2015. PloS one.
10- Goldblatt, Ran, & et al. (2016). Detecting the boundaries of urban areas in india: A dataset for pixel-based image classification in google earth engine. Remote Sensing.
11- Gorelick, Noel, & et al. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote sensing of Environment.
12- Grippa, T., Georganos, S., Zarougui, S., Bognounou, P., Diboulo, E., Forget, Y., ... & Wolff, E. (2018).
Mapping urban land use at street block level using openstreetmap, remote sensing data, and spatial metrics. ISPRS International Journal of Geo-Information, 7(7), 246.
13- He, Chunyang, & et al. (2010). Improving the normalized difference built-up index to map urban built-up areas using a semiautomatic segmentation approach. Remote Sensing Letters, 1.4, 213-221.
14- Hsu, Chih-Wei, Chih-Chung Chang, & Chih-Jen Lin. (2003). A practical guide to support vector classification.
15- Kaynarca, Mustafa, & Nusret Demir. (2018). Detection of Urban Buildings by Using Multispectral Gokturk-2 and Sentinel 1A Synthetic Aperture Radar Images. Multidisciplinary Digital Publishing Institute Proceedings., 2.
16- Li, Lu, & et al. (2019). Residual Unet for Urban Building Change Detection with Sentinel-1 SAR Data. IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE.
17- Liu, X., Hu, G., Chen, Y., Li, X., Xu, X., Li, S., ... & Wang, S. (2018). High-resolution multi-temporal mapping
of global urban land using Landsat images based on the Google Earth Engine Platform. Remote sensing of environment, 209, 227-239
18- Li, Q., Qiu, C., Ma, L., Schmitt, M., & Zhu, X. X. (2020). Mapping the land cover of Africa at 10 m resolution
from multi-source remote sensing data with Google Earth Engine. Remote Sensing, 12(4), 602.
19- Luo, N., Wan, T., Hao, H., & Lu, Q. (2019). Fusing high-spatial-resolution remotely sensed imagery and
OpenStreetMap data for land cover classification over urban areas. Remote Sensing, 11(1), 88.
20- Mahalanobis, & Chandra, P. (1936). On the generalized distance in statistics. National Institute of Science of India.
21- Martone, Michele, & et al. (2018). The global forest/non-forest map from TanDEM-X interferometric SAR data. Remote sensing of environment.
22- Otsu, & Nobuyuki. (1979). A threshold selection method from gray-level histograms. IEEE transactions on systems, man, and cybernetics.
23- Pal, & Mahesh. (2005). Random forest classifier for remote sensing classification. International Journal of Remote Sensing.
24- Pesaresi, Martino, & et al. (2016). Assessment of the added-value of Sentinel-2 for detecting built-up areas. Remote Sensing.
25- Potere, David, & et al. (2009). Mapping urban areas on a global scale: which of the eight maps now available is more accurate? International Journal of Remote Sensing.
26- Protopapadakis, E., Doulamis, A., Doulamis, N., & Maltezos, E. (2021). Stacked autoencoders driven by semi-
supervised learning for building extraction from near infrared remote sensing imagery. Remote Sensing, 13(3), 371.
27- Qiu, Chunping, & et al. (2020). A framework for large-scale mapping of human settlement extent from Sentinel-2 images via fully convolutional neural networks. ISPRS Journal of Photogrammetry and Remote Sensing, 163, 152-170.
28- Richards, John A, & J. A. Richards. (1999). Remote sensing digital image analysis. Springer.
29- Sarvestani, Mahdi Sabet, Ab Latif Ibrahim, & Pavlos Kanaroglou. (2011). Three decades of urban growth in the city of Shiraz, Iran: A remote sensing and geographic information systems application. Cities.
30- Schneider, & et al. (2010). Mapping global urban areas using MODIS 500-m data: New methods and datasets based on ‘urban ecoregions’. Remote Sensing of Environment.
31- Tucker, & Compton J. (1978). Red and photographic infrared linear combinations for monitoring vegetation. Remote Sensing OF Environment.
32- Van Zyl, & Jakob J. (2001). The Shuttle Radar Topography Mission (SRTM): a breakthrough in remote sensing of topography. Acta Astronautica.
33- Xiao, Yinghui, & Qingming Zhan. (2009). A review of remote sensing applications in urban planning and management in China. 2009 Joint Urban Remote Sensing Event, IEEE(IEEE).
34- Zakeri, H., Fumio Yamazaki, & Wen Liu. (2017). Texture analysis and land cover classification of Tehran using polarimetric synthetic aperture radar imagery. Applied Sciences.
35- Zha, Yong, Jay Gao, & Shaoxiang Ni. (2003). Use of normalized difference built-up index in automatically mapping urban areas from TM imagery. International journal of remote sensing.
36- Zong, L., He, S., Lian, J., Bie, Q., Wang, X., Dong, J., & Xie, Y. (2020). Detailed Mapping of Urban Land Use
Based on Multi-Source Data: A Case Study of Lanzhou. Remote Sensing, 12(12), 1987.