شناسایی و تفکیک پوشش های زمین با استفاده از تلفیق تصاویر اپتیک و رادار

محورهای موضوعی : منابع طبیعی و مدیریت زیست محیطی

مصطفی کابلی زاده ^{1 *} , سجاد زارعی ² , رحمان خنافره ³

1 - گروه سنجش از دور و GIS، دانشکده علوم زمین، دانشگاه شهید چمران اهواز
2 - گروه سنجش از دور و GIS، دانشکده علوم زمین؛ دانشگاه شهید چمران اهواز
3 - گروه سنجش از دور و GIS، دانشکده علوم زمین؛ دانشگاه شهید چمران اهواز

تاریخ دریافت : 1401/10/03 تاریخ پذیرش : 1401/11/08 تاریخ انتشار : 1404/04/31

کلید واژه: انتخاب ویژگی, پوشش‌های زمین در محیط‌های شهری, روش ماشین بردار پشتیبان (SVM), طبقه‌بندی, ویژگیهای بافتی,

چکیده مقاله :

طبقه‌بندی و تفکیک پوشش زمین از مهم‌ترین کاربردهای سنجش از دور می‌باشد. برای انجام طبقه‌بندی، داده‌های ماهوارهای چندطیفی ابزاری کارآمد می‌باشند، اما متاسفانه در برخی از شرایط، مانند آب و هوای ابری در دسترس نیستند. همچنین اکثر الگوریتم‌های طبقه‌بندی داده‌های سنجش از دور بر اساس ویژگی‌ها و اطلاعات طیفی پیکسل‌ها عمل می‌کنند که این مسئله باعث نادیده گرفتن اطلاعات مکانی مفید قابل استخراج از تصاویر، از جمله؛ بافت تصاویر می‌شود. استفاده همزمان از بافت و اطلاعات طیفی مبحثی است که به آن کمتر پرداخته شده است. بنابراین با در نظر گرفتن این ایده جهت انتخاب ویژگی‌های بهینه برای تهیه نقشه پوشش‌های زمین از دو روش استفاده شد. روش اول بازتاب نرمال شده عوارض با توجه به ویژگی‌های استخراج شده و روش دوم اعمال شاخص ضریب بهینه (Optimum Index Factor) بروی ویژگی-های بافتی و طیفی استخراج شده می‌باشد. به این منظور فرآیند طبقه‌بندی با استفاده از روش ماشین بردار (Support Vector Machine)، برروی تصویر راداری سنتینل-1و تصویر چندطیفی سنتینل- 2، ویژگی‌های بهینه انتخاب شده با دو روش و ترکیب باندهای تصویر با ویژگی‌های بهینه انتخاب شده با دو روش و در آخر با تلفیق بهترین ترکیب باندهای رادار و اپتیک انجام گرفت. با توجه به نتایج بدست آمده، طبقه‌بندی با استفاده از ویژگی‌های طیفی دقت بالاتری نسبت به طبقه‌‌بندی با استفاده از ویژگی‌های بافت دارد. با تلفیق ویژگی‌های اپتیک و رادار و بدست آمدن مقادیر 07/97 درصد برای دقت کلی و 96/0 برای ضریب کاپا دقت طبقه بندی تا حد زیادی بهبود داده شد. این تحقیق نشان داد که با انتخاب ویژگی‌های بهینه و تلفیق داده‌های طیفی و راداری می‌توان از ویژگی‌های متفاوت هر یک از داده‌ها استفاده کرد و به نتایج بهتری رسید. همچنین تلفیق ویژگی بافتی از تصویر راداری و ویژگی طیفی از تصویر اپتیکی می‌تواند تاثیر بسیار خوبی در بهبود نتایج طبقه‌بندی پوشش زمین داشته باشد.

چکیده انگلیسی:

Classification and separation of land cover is one of the most important applications of remote sensing. To perform classification, multispectral satellite data are an efficient tool, but unfortunately, they are not available in some conditions, such as cloudy weather. Also, most remote sensing data classification algorithms operate based on the characteristics and spectral information of pixels, which causes the useful spatial information that can be extracted from the images to be ignored, including; The texture of the pictures. The simultaneous use of texture and spectral information is a topic that has been less discussed. Therefore, considering this idea, two methods were used to select the optimal features for preparing the land cover map. The first method is the normalized reflection of complications according to the extracted features and the second method is applying the Optimum Index Factor on the extracted textural and spectral features. For this purpose, the classification process using the Support Vector Machine method, on the Sentinel-1 radar image and the Sentinel-2 multispectral image, the optimal features selected by the two methods and the combination of image bands with the optimal features selected by It was done by two methods and finally by combining the best combination of radar and optical bands. According to the obtained results, the classification using spectral features is more accurate than the classification using texture features. By combining the optical and radar features and obtaining values of 97.07% for the overall accuracy and 0.96% for the Kappa coefficient, the classification accuracy was improved to a great extent. This research showed that by choosing optimal features and combining spectral and radar data, different features of each data can be used and better results can be achieved.

منابع و مأخذ:

Albregtsen F. 2008. Statistical texture measures computed from gray level coocurrence matrices. Image processing laboratory, department of informatics, university of oslo.
Clerici N, Valbuena Calderón C, Posada J. 2017. Fusion of Sentinel-1A and Sentinel-2A data for land cover mapping: a case study in the lower Magdalena region, Colombia. Journal of Maps, 13(2), 718-726.
Cervantes J, Garcia-Lamont F, Rodríguez-Mazahua L, Lopez A. 2020. A comprehensive survey on support vector machine classification: Applications, challenges and trends. Neurocomputing, 408: 189-215
Congalton RG, Green K. 2019. Assessing the accuracy of remotely sensed data: principles and practices. CRC press.
Dihkan M, Guneroglu N, Karsli F, Guneroglu A. 2013. Remote sensing of tea plantations using an SVM classifier and pattern-based accuracy assessment technique. International Journal of Remote Sensing, 34(23), 8549-8565.
Filipponi F. 2019. Sentinel-1 GRD preprocessing workflow. Multidisciplinary digital publishing institute proceedings, 18(1), 11.
Haas J, Ban Y. 2017. Sentinel-1A SAR and sentinel-2A MSI data fusion for urban ecosystem service mapping. Remote Sensing Applications: Society and Environment, 8, 41-53.
Hu J, Ghamisi P, Zhu X. 2018. Feature extraction and selection of sentinel-1 dual-pol data for global-scale local climate zone classification. ISPRS International Journal of Geo-Information, 7(9), 379.
Haghighat M, Zonouz S, Abdel-Mottaleb M. 2013. Identification using encrypted biometrics. In International Conference on Computer Analysis of Images and Patterns (pp. 440-448). Springer, Berlin, Heidelberg.
Hoseinpour S, Mohammadzadeh A, Eslami M. 2017. Evaluation of the Textural Statistics of the Gray Level Co-Occurrence Matrix Performance for Change Detection. Journal of Geomatics Science and Technology, 6(3), 119-129.
Han, S, Li H, Gu H. 2008. The study on image fusion for high spatial resolution remote sensing images. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. XXXVII. Part B, 7, 1159-1164.
Ling J, Zhang H, Lin Y. 2021. Improving Urban Land Cover Classification in Cloud-Prone Areas with Polarimetric SAR Images. Remote Sensing, 13(22), 4708.
Ienco D, Interdonato R, Gaetano R, Minh D. H. T. 2019. Combining Sentinel-1 and Sentinel-2 Satellite Image Time Series for land cover mapping via a multi-source deep learning architecture. ISPRS Journal of Photogrammetry and Remote Sensing, 158, 11-22.
Joshi N, Baumann M, Ehammer A, Fensholt R, Grogan K, Hostert P, Waske B. 2016. A review of the application of optical and radar remote sensing data fusion to land use mapping and monitoring. Remote Sensing, 8(1), 70.
Jafar Biglou M, Saman N, Kamyar E, Fatmeh M. 2018. Investigation and prediction of land use changes for lands under the influence of sand dunes in the eastern of the city of Boshrooyeh using Remote Sensing and Geographic Information System, 6(3), 117-134. (In Persian)
Jalby W, Philippe B. 1991. Stability analysis and improvement of the block Gram–Schmidt algorithm. SIAM journal on scientific and statistical computing, 12(5), 1058-1073.
Kabulizadeh M, Rangzan K, Mohammadi Sh. 2017. Application of combining Landsat-8 and Sentinel-2 satellite images in environmental monitoring. Remote Sensing and Geographical Information System in Natural Resources, 9(3), 53-71. (In Persian)
Kranjčić N, Medak D, Župan R, Rezo M. 2019. Support vector machine accuracy assessment for extracting green urban areas in towns. Remote Sensing, 11(6), 655.
Kim C. 2016. Land use classification and land use change analysis using satellite images in Lombok Island, Indonesia. Forest Science and Technology, 12(4), 183-191.
Kong Y, Hong F, Leung H, Peng X. 2021. A Fusion Method of Optical Image and SAR Image Based on Dense-UGAN and Gram–Schmidt Transformation. Remote Sensing, 13(21), 4274.
Lee J, Grunes M, Ainsworth T, Du L, Schuler, D, Cloude S. 1999. Unsupervised classification using polarimetric decomposition and the complex Wishart classifier. IEEE Transactions on Geoscience and Remote Sensing, 37(5), 2249-2258.
Maleknejad Yazdi A, Ghasemian H, Cesei V, Shahesvari A, Kosha H. 2014. Investigating the improvement of classification accuracy using single-band ALI image fusion with Hyperion hyperspectral images. Iran Remote Sensing and GIS, 7(3), 65-82. (In Persian)
Mohammad Nejad Niazi S, Mokhtar Zade M, Saeed Zadeh F. 2016. A Novel IHS-GA Fusion Method Based on Enhancement Vegetated Area. JGST, Journal of Geomatics Sciences and Techniques, 6 (1), 235-248. (In Persian)
Mathur A, Foody G. 2008. Multiclass and binary SVM classification: Implications for training and classification users. IEEE Geoscience and remote sensing letters, 5(2), 241-245.
Mavroforakis M, Theodoridis S. 2006. A geometric approach to support vector machine (SVM) classification. IEEE transactions on neural networks, 17(3), 671-682.
Nadizadeh Shorabeh S, Attarchi S, Minaei, F. 2020. Evaluation of the Capability of Polarimetric Radar Bands to Extract Biophysical Properties of the Earth's Surface. Physical Geography Research Quarterly, 52(1), 147-164. (In Persian)
Nadizadeh Shorabeh S, Hamzeh S, Kiavarz M, Afsharipoor S. 2018. Effects of Spatial and Temporal Land Use Changes and Urban Development on the Increase of Land Surface Temperature Using Landsat Multi-Temporal Images (Case study: Gorgan City). Geographical Urban Planning Research (GUPR), 6(3), 545-568. (In Persian)
Orynbaikyzy A, Gessner U, Conrad C. 2019. Crop type classification using a combination of optical and radar remote sensing data: A review. International journal of remote sensing, 40(17), 6553-6595.
Palubinskas G, Reinartz P, Bamler R. 2010. Image acquisition geometry analysis for the fusion of optical and radar remote sensing data. International Journal of Image and Data Fusion, 1(3), 271-282.
Pereira L, D O, Freitas C, Sant´ Anna S, Lu, D, Moran E. 2013. Optical and radar data integration for land use and land cover mapping in the Brazilian Amazon. GIScience , remote sensing, 50(3), 301-321.
Potin P, Rosich B, Grimont P, Miranda N, Shurmer I, O'Connell A, Krassenburg M. 2016. Sentinel-1 mission status. In Proceedings of EUSAR 2016: 11th European Conference on Synthetic Aperture Radar (pp. 1-6). VDE.
Rangzan K, Kabolizadeh M, Karimi D, Zareie S. 2019. Supervised cross-fusion method: a new triplet approach to fuse thermal, radar, and optical satellite data for land use classification. Environmental monitoring and assessment, 191(8), 1-12.
Sicre C. M, Fieuzal R, Baup F. 2020. Contribution of multispectral (optical and radar) satellite images to the classification of agricultural surfaces. International Journal of Applied Earth Observation and Geoinformation, 84, 101972.
Shao Z, Wu W, Guo S. 2020. IHS-GTF: A fusion method for optical and synthetic aperture radar data. Remote Sensing, 12(17), 2796.
Sarzynski T, Giam X, Carrasco L, Lee J. 2020. Combining radar and optical imagery to map oil palm plantations in Sumatra, Indonesia, using the Google Earth Engine. Remote Sensing, 12(7), 1220.
Shakriyari M, Ehsani A, Afsari R. 2018. Determination of optimal method of classification and mapping of land cover with a comparison of artificial neural network and support vector machine algorithms by satellite data Case study: (international Hamoun wetland). Journal of Environmental Science and Technology, 4(79), 193-208. (In Persian)
Saraskanrood S, khodabandelo B, Naseri A, moradi A. 2019. Extracting Land Use Map based on a comparison between Pixel-Based and Object-Oriented Classification Methods Case Study: Zanjan City, Scientific- Research Quarterly of Geographical Data (SEPEHR), 28(110), 195-208. (In Persian)
Teimouri M, Mokhtarzade M, Valadan Z. 2017. Building Detection in Urban Areas using Features Fusion of Optical and Radar Images in Neural Networks, Journal of Geospatial information technology, 4 (4), 33-51. (In Persian)
Tewabe D, Fentahun T. 2020. Assessing land use and land cover change detection using remote sensing in the Lake Tana Basin, Northwest Ethiopia. Cogent Environmental Science, 6(1), 1778998.
Tao Z, Jun L, Keming Y, Wenshan L, Yuyu Z. 2015. Fusion algorithm for hyperspectral remote sensing image combined with harmonic analysis and gram-schmidt transform. Acta Geodaetica et Cartographica Sinica, 44(9), 1042.18(1), 11.
Wu R, Liu G, Zhang R, Wang X, Li Y, Zhang B, Xiang, W. 2020. A deep learning method for mapping glacial lakes
Wang L, Jia Y, Yao Y, Xu D. 2019. Accuracy Assessment of Land Use Classification Using Support Vector Machine and Neural Network for Coal Mining Area of Hegang City, China. Nature Environment , Pollution Technology, 18(1).
Yue S, Li P, Hao P. 2003. SVM classification: Its contents and challenges. Applied Mathematics-A Journal of Chinese Universities, 18(3), 332-342.
Yu H, Kim S. 2012. SVM Tutorial-Classification, Regression and Ranking. Handbook of Natural computing, 1, 479-506.
Yilmaz V, Serifoglu Yilmaz C, Güngör O, Shan J. 2020. A genetic algorithm solution to the gram-schmidt image fusion. International Journal of Remote Sensing, 41(4), 1458-1485.
Zhang H, Li J, Wang T, Lin H, Zheng Z, Li Y, Lu Y. 2018. A manifold learning approach to urban land cover classification with optical and radar data. Landscape and Urban Planning, 172, 11-24.