ارائه روشی تلفیقی مبتنی بر الگوریتم‌های طبقه‌بندی پارامتریک و غیر پارامتریک به منظور جداسازی پوشش‌های مختلف در جنگل‌های هیرکانی

نوع مقاله : علمی - پژوهشی

نویسندگان

دانشکده مهندسی نقشه برداری و اطلاعات مکانی، پردیس دانشکده‌های فنی، دانشگاه تهران

چکیده

تهیه نقشه‌های پوشش اراضی با دقت بالا، همواره یکی از اهداف مهم محققان در زمینه مدیریت اراضی بوده است. هدف از‌این پژوهش، ارائه روش نوینی جهت تهیه نقشه‌های کاربری اراضی با استفاده از پردازش تصاویر ماهواره‌ای بوده است. به همین منظور، از تصاویر ماهواره لندست 8، به عنوان تصویر پایه و نقشه مدل رقومی ارتفاعی (DEM)، داده‌های حاصل از تجزیه به عنوان مولفه‌­های اصلی و شاخص‌های طیفی جهت استخراج نقشه پوشش اراضی در منطقه مطالعاتی استفاده شد. پس از پیش‌پردازش‌ها و آماده‌سازی داده‌های مورد نیاز، اقدام به تهیه نمونه‌های آموزشی شد. در‌این پژوهش، نمونه‌های آموزشی در دو بخش به کار گرفته شدند؛ در بخش اول از آنها به عنوان ورودی، جهت طبقه‌بندی تصویر با الگوریتم‌های نظارت شده، حداکثر احتمال و ماشین بردار پشتیبان استفاده شد. در بخش دوم، به‌منظور طبقه‌بندی با روش درخت تصمیم گیری، از‌ این نمونه‌ها برای تعیین محدوده بازتاب طیفی هر پوشش در طیف امواج الکترومغناطیس (باندهای تصویر، PCA، شاخص‌های طیفی و DEM) استفاده شد. سپس با استفاده از ‌این داده‌ها و شروط دودویی درخت تصمیم‌گیری، هر پوشش مشخص و نقشه پوشش آن استخراج شد. پس از تهیه نقشه‌های ذکر شده، به منظور تلفیق نتایج طبقه‌بندی و حصول دقت بالاتر، از روش حداکثر رای‌گیری به منظور تهیه نقشه تلفیقی جدید پوشش اراضی منطقه استفاده شد. همچنین به منظور ارزیابی دقت نقشه‌های تولیدی، از پارامترهای آماری منتج از ماتریس ابهام شامل دقت کلی، ضریب کاپا، دقت کاربر و دقت تولیدکننده استفاده شد. بر اساس نتایج حاصله، روش تلفیقی با دقت کلی 37/93 درصد و ضریب کاپا 91/0 دارای بیشترین دقت بوده است. دقت کلی نقشه پوشش روش درخت تصمیم‌گیری، ماشین بردار پشتیبان و حداکثر احتمال نیز به ترتیب 61/89، 01/88 و 6/87 درصد بوده‌اند. با توجه به‌اینکه در طبیعت پوشش خالص، به ندرت مشاهده می‌شود و بیشتر پوشش‌ها به صورت ترکیبی وجود دارند، لذا بهتر است از روش‌های نوینی که همه ابعاد پدیده‌ها را پوشش می‌دهند استفاده شود. در‌این پژوهش، اطلاعات حاصل از طبقه‌بندی نظارت شده و همچنین اطلاعات حاصل از روش منطقی درخت تصمیم‌گیری با یکدیگر تلفیق شده و نتایج حاصله به خوبی، بیانگر بهبود دقت نهایی طبقه‌بندی بودند.

کلیدواژه‌ها


عنوان مقاله [English]

A hybrid classification method based on fusion of parametric and non-parametric classification algorithms for Landuse/Landcover map in Hirkani Forests

نویسندگان [English]

  • Mohammad Saadat
  • Reza Shahhoseini
College of Engineering, University of Tehran, Tehran, Iran
چکیده [English]

Preparation of proper land use maps has always been one of the important goals of researchers and policymakers. The aim of this study was to provide a new method for preparing land use maps using remotely sensed data and satellite data imagery. For this Purpose, we used Landsat 8 data, Digital Elevation Model (DEM), Principal Component Analysis (PCA), and Spectral Indices to extract land use map in the study area. After all required preprocessing, the training samples were provided. In this study, the training samples were utilized in two parts; in the first part they were used as inputs for image classification using supervised algorithms of maximum likelihood Classification (MLC) and support vector machine (SVM). In the second part, in order to applying Decision Tree Classification (DTC), these training samples were used to determine the spectral reflection of each end-member in the spectrum of electromagnetic waves (image bands, PCA, spectral indices, and DEM).Then, using these binary data and DTC, each end-member was identified and the Landuse/Landcover (LULC) map was extracted. In order to combine the classification results and achieve higher accuracy, the Majority Vote Classification (MVC) method was applied to prepare a new compilation of land use in the area. In order to evaluate the accuracy of produced maps, the statistical parameters extracted from the confusion matrix including overall accuracy, kappa coefficient, user and producer’s accuracy were utilized. According to the results, the combined method (MVC) with a total accuracy of 93.37% and kappa coefficient of 0.91 had the highest accuracy. The overall accuracy of the DTC, SVM, and MLC were 89.61, 88.01 and 87.6%, respectively. Due to the fact that in the nature most of the landuse are mixed and complicated, it would be better to use new methods that cover all aspects of the phenomena. In this research, the data extracted from the supervised classifications as well as the data derived from the DTC were combined and the results clearly illustrate the improvement of the final accuracy of the classification.

کلیدواژه‌ها [English]

  • Majority Vote Classification
  • Landuse
  • Classification Algorithm
  • Remote Sensing
  • Landsat 8
  1. Salajegheh, A., Razavizade, S., Khorasani, N., Hamidifar, M., & Salajegheh, S. 2011, Land Use Changes and its Effects on River Water Quality (Case Study: Karkheh Watershed), Journal of Environmental Studies, Vol. 58, pp. 81-86.
  2. Helming, K., Perez-Soba, M., & Tabbush, P., 2008, Sustainability Impact Assessment of Land Use Changes.USA: Springer-Verlag Berlin Heidelber Press.
  3. Amirnejad, H., 2013, Factors Affecting Farmers' Willingness to Changing Land Use in Mazandaran Province, Journal of Agricultural Economics Reasearch, Vol. 5(4), pp. 87-106.
  4. Li, Z., Li, X., Wang, Y., Ma, A. & Wang, J., 2004, Land-use change analysis in Yulin prefecture, northwestern China using remote sensing and GIS, International Journal of Remote Senseing, Vol. 51, pp. 23-55,.
  5. Shi, Y., Wang, R., Fan, L., Li, J., & Yang, D., 2010, Analysis on Land-use Change and Its Demographic Factors in the Original-stream Watershed of Tarim River Based on GIS and Statistic”, Procedia Environmental Sciences, Vol. 2, pp. 175-184,.
  6. Al Rawashdeh, S. B., 2012, Assessment of Change Detection Method Based on Normalized Vegetation Index in Environmental Studies, International Journal of Applied Science and Engineering, Vol. 10(2), pp. 89-97,.
  7. Sabzghabaei, Gh., Dashti, S., Bazm Ara Baleshti, M. & Jafarzadeh K., 2015, Detecting the variability process of the protected area of Hara Khourkhouran, Journal of Marine Biology, Vol. 7(26), pp. 1-12,.
  8. Sanhouse-Garcia, A. J., Rangel-Peraza, J. G., Bustos-Terrones, Y., Garcia-Ferrer, A. & Mesas Carrascosa, F. J., 2016, Land Use Mapping from CBERS-2 Images with Open Source Tools by Applying Different Classification Algorithms, Physics and Chemistry of the Earth, Parts A/B/C, Vol. 91, pp. 27-37,.
  9. Shao, Y. & Lunetta, R.S., 2012, Comparison of Support Vector Machine, Neural Network, and CART Algorithms for the Land-Cover Classification Using Limited Training Data Points, ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 70, pp. 78–87.
  10. Mantero, P., Moser, G., and Serpico, S. B., 2005, Partially supervised classification of remote sensing images through SVM-based probability density estimation, IEEE Transactions on Geoscience and Remote Sensing, Vol. 43(3), pp. 559–570,.
  11. Li, C., Wang, j., Wang, L., Hu, L. & Gong, P., 2014, Comparison of Classification Algorithms and Training Sample Sizes in Urban Land Classification with Landsat Thematic Mapper Imagery, Remote Sensing, Vol. 6 (2), pp. 964-983,.
  12. Alizadeh, M., Mirzaie, R. & Kia, H., 2016, Comparative study of multiple classification methods for land use mapping (Case study: Kann and Karaj rivers basin), Geography and Sustainability of Environment, Vol. 20, pp. 89-103,.
  13. Zandieh, V. 2015, Assessment of land use changes and its impact on groundwater resources based on digital images and GIS analysis in Malayer plain, Hamedan province, Master thesis, Department of Soil Science at Lorestan University,.
  14. Yousofi, S., Tazeh, M., Mirzaie, S. & Tavangar, Sh., 2014, Comparison of different satellite image classification algorithms for land use mapping (case study: Noor city), Journal of RS and GIS for Natural Resources, Vol. 5(3), pp. 67-76,.
  15. Fathizad, H., Tazeh, M. & Kalantari, S. 2015, Comparison of the Efficiency of Pixel based Classification Methods (Fuzzy, Artmap Fuzzy, Neural Networks and Tree Decision) and Object based Methods for Land Use Mapping (Case Study: Dry and Semi-Dry Watershed of Meymeh, Ilam Province), Journal of Khoshkboom, Vol. 5(2), pp. 69-82,.
  16. Jamil, A. and Bayram, B. 2018, Tree Species Extraction and Land Use/Cover Classification from High-Resolution Digital Orthophoto Maps, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 11(1), pp. 89-94,.
  17. Kittler, J. and Alkoot, F. M., 2003, Sum versus vote fusion in multiple classifier systems, IEEE transactions on pattern analysis and machine intelligence, Vol. 25(1), pp. 110-115,.
  18. James, G. 1998, Majority vote classifiers: theory and applications (Doctoral disserta-tion, Stanford University).
  19. Abdul Qadir, A., Med-hut, B., & Jirjees, Th., 2010, Monitoring and evaluation of soil salinity in term of spectral response using landsat images and GIS in Mesopotamian plain/Iraq, Journal of Iraqi Desert Studies, Vol. 2(2), pp. 19-32,.
  20. Fatemi, B. & Rezaie, Y., 2012, Fundamental of Remote Sensing. Iran: Azadeh Press,.
  21. Ren, H., Zhou, G. & Zhang, F., 2018, Using negative soil adjustment factor in soil-adjusted vegetation index (SAVI) for aboveground living biomass estimation in arid grasslands, Remote Sensing of Environment, Vol. 209, pp. 439-445,.
  22. Hasanloo, M. & Samadzadegan, F., 2013, Estimation of inherent dimension in Hyperspectral satellite imeges, Journal of Geomatics Science and Technology, Vol. 3(3), pp. 101-109,.
  23. Otukei, J. R. & Blaschke, T., 2010, Land cover change assessment using decision trees, support vector machines and maximum likelihood classification algorithms, International Journal of Applied Earth Observation and Geoinformation, Vol. 12, pp. 27-31,.
  24. Mountrakis, G., Im, J., & Ogole, C., 2011, Support vector machines in remote sensing: A review ISPRS, Journal of Photogrammetry and Remote Sensing, Vol. 66(3), pp. 247-259,.
  25. Vapnik, W. N., 1999, An overview of statistical learning theory, IEEE Transactions of Neural Networks, Vol. 10, pp. 988–999,.
  26. Cortes, C. and Vapnik, V., 1995, Support-vector networks, Machine Learning, Vol. 20(3), pp. 273–297,.
  27. Zhu, G. and Blumberg, D. G., 2002, Classification using ASTER data and SVM algorithms; The case study of Beer Sheva, Israel, Remote Sensing of Environment, Vol. 80(2), pp. 233–240,.
  28. Mather, P. and Tso, B., 2016, Classification methods for remotely sensed data, CRC press,.
  29. Hare, S., Golodetz, S., Saffari, A., Vineet, V., Cheng, M. M., Hicks, S. L. & P. H. Torr, 2016, Struck: Structured output tracking with kernels, IEEE transactions on pattern analysis and machine intelligence, Vol. 38(10), pp. 2096-2109,.
  30. Tang, Y., 2013, Deep learning using linear support vector machines, arXiv preprint arXiv:1306.0239,.
  31. Zhu, G. B., Liu, X. L. & Jia, Z. G., 2006, A multi-resolution hierarchy classification study compared with conservative methods, ISPRSWG II/3, II/6Workshop Multiple representation and interoperability of spatial data. Hanover, Germany,.
  32. Chasmer, L., Hopkinson, C., Veness, T., Quinton, W., & Baltzer, J., 2014, A decision-tree classification for low-lying complex land cover types within the zone of discontinuous permafrost, Remote Sensing of Environment, Vol. 143, pp. 73-84,.
  33. Rouse, Jr. J., Haas, R. H., Schell, J. A., & Deering, D. W., 1974, Monitoring vegetation systems in the Great Plains with ERTS.
  34. Khan, S. I., Hong, Y., Wang, J., Yilmaz, K. K., Gourley, J. J., Adler, R. F., & Irwin, D., 2011, Satellite remote sensing and hydrologic modeling for flood inundation mapping in Lake Victoria basin: Implications for hydrologic prediction in ungauged basins, IEEE Transactions on Geoscience and Remote Sensing, Vol. 49(1), pp. 85-95,.
  35. Mather, M. M., 1999, Computer processing of remotely sensed data, 2nd Edition, John Wiley & Sons Press.
  36. Arkhi, S., 2012, Assessment of the Effectiveness of Decision Tree Classification Method for Extracting Land uses Map by Using Satellite Data in Cham Gardalan Catchment Area, Geography and Territorial Spatial Arrangement, Vol. 4, pp. 17-26,.