قابليت الگوريتم‌هاي نظارت شده در تهيه نقشه پوشش اراضي در مقياس محلي (مطالعه موردي: استان گيلان

زﻣﯿﻨﻪ و اﻫﺪاف: اﻣﮑﺎن ﺑﺮرﺳﯽ ﭘﻮﺷﺶ زﻣﯿﻦ در ﻣﻘﯿﺎس ﮔﺴﺘﺮده ﺑﺎ اﺳﺘﻔﺎده از دادهﻫﺎي ﺳﻨﺠﺶ از دور وﺟﻮد دارد. ﻃﺒﻘﻪﺑﻨﺪي ﭘﻮﺷﺶ زﻣﯿﻦ در اﺳﺘﺎن ﮔﯿﻼن ﺑﺎ اﺳﺘﻔﺎده از ﺳﻨﺠﻨﺪه OLI و 4 ﮐﺮﻧﻞ ﻣﺎﺷﯿﻦﺑﺮدار ﭘﺸﺘﯿﺒﺎن )SVM(، ﺷﺒﮑﻪ ﻋﺼﺒﯽ ﻣﺼﻨﻮﻋﯽ )ANN( و ﺣﺪاﮐﺜﺮ اﺣﺘﻤﺎل )ML( اﻧﺠﺎم ﺷﺪ. روش ﺑﺮرﺳﯽ: ﻃﺒﻘﻪﺑﻨﺪيﻫﺎ ﺑﺮ اﺳﺎس ﻧﻤﻮﻧﻪﻫﺎي ﺗﻌﻠﯿﻤﯽ 10 ﭘﻮﺷﺶ ﻣﺨﺘﻠﻒ در ﮐﻞ اﺳﺘﺎن ﺻﻮرت ﮔﺮﻓﺖ. ﺑﺮاي ﺑﺎﻻﺑﺮدن دﻗﺖ ﻧﻘﺸﻪﻫﺎ، ﺗﺼﻮﯾﺮ OLI ﺑﺎ اﺳﺘﻔﺎده از ﻣﺤﺼﻮﻻت MODIS ﺑﺎ اﻋﻤﺎل ﮐﺪ اﻧﺘﻘﺎل ﺗﺎﺑﺸﯽ وﮐﺘﻮري در ﻃﯿﻒ ﺧﻮرﺷﯿﺪ )6SV( ﻣﻮرد ﺗﺼﺤﯿﺢ اﺗﻤﺴﻔﺮي ﻗﺮار ﮔﺮﻓﺘﻪ اﺳﺖ. ﺗﺼﻮﯾﺮ ﺑﺮ ﻣﺒﻨﺎي ﻣﻌﯿﺎر ﻫﻤﮕﻨﯽ ﺑﻪ 219000 ﭘﻠﯽﮔﻮن، ﺳﮕﻤﻨﺖﺑﻨﺪي ﮔﺮدﯾﺪ. ﺑﻪ روش ﮐﺎﻣﻼَ ﺗﺼﺎدﻓﯽ 2% از ﭘﻠﯽﮔﻮنﻫﺎي ﻫﻤﮕﻦ ﺑﺮاي آﻣﻮزش و آزﻣﻮن اﺳﺘﻔﺎده ﮔﺮدﯾﺪ. ﺑﺎ ﺑﺎزدﯾﺪ ﻣﯿﺪاﻧﯽ، ﭘﻠﯽﮔﻮنﻫﺎ ﺑﻪ ﮐﻼسﻫﺎ ﺑﺮﭼﺴﺐ داده ﺷﺪﻧﺪ. ﯾﺎﻓﺘﻪﻫﺎ: ﺑﻪ ﮐﺎرﮔﯿﺮي ﺗﺼﺎوﯾﺮ ﺗﺼﺤﯿﺢ ﺷﺪه ﺑﺎ ﮐﺪ 6SV در ﻃﺒﻘﻪﺑﻨﺪي ﺳﺒﺐ ارﺗﻘﺎء ﺻﺤﺖ ﮐﻠﯽ اﻟﮕﻮرﯾﺘﻢﻫﺎي SVM ،ANN و ML ﺑﻪ ﺗﺮﺗﯿﺐ ﺑﻪ ﻣﯿﺰان 0/8 ،%0/11% و 1/9% ﮔﺮدﯾﺪه اﺳﺖ. ارزﯾﺎﺑﯽ ﻧﺘﺎﯾﺞ ﺑﯿﺎنﮔﺮ ﺑﺮﺗﺮي ﮐﺮﻧﻞ ﺷﻌﺎﻋﯽ SVM ﺑﻪ ﺗﺮﺗﯿﺐ ﺑﺎ ﺻﺤﺖ ﮐﻠﯽ و ﺿﺮﯾﺐ ﮐﺎﭘﺎي آﻣﺎري 75/6% و 0/72 اﺳﺖ. در اﯾﻦ اﻟﮕﻮرﯾﺘﻢ ﺻﺤﺖ ﮐﻼسﻫﺎي ﮐﺸﺎورزي، ﻣﺮاﺗﻊ ﻣﺸﺠﺮ و آﺑﯽ ﺑﻪ ﺗﺮﺗﯿﺐ 72/55 ،%93/16% و 96/57% اﺳﺖ. ﻧﺘﺎﯾﺞ ﺑﯿﺎنﮔﺮ ارﺗﻘﺎء ﺻﺤﺖ ﮐﻠﯽ اﻟﮕﻮرﯾﺘﻢ SVM در ﻣﻘﺎﯾﺴﻪ ﺑﺎ اﻟﮕﻮرﯾﺘﻢ ML ﺑﻪ ﻣﯿﺰان 1/67% اﺳﺖ. ﺑﺤﺚ و ﻧﺘﯿﺠﻪﮔﯿﺮي: اﯾﻦ ﺗﺤﻘﯿﻖ ﻧﺸﺎندﻫﻨﺪه ﺑﺮﺗﺮي روش ﻧﺎﭘﺎراﻣﺘﺮﯾﮏ SVM در ﻣﻘﺎﯾﺴﻪ ﺑﺎ ﭘﺎراﻣﺘﺮﯾﮏ در ﺗﻬﯿﻪ ﻧﻘﺸﻪ ﭘﻮﺷﺶ اراﺿﯽ اﺳﺘﺎن ﮔﯿﻼن اﺳﺖ. اﻋﻤﺎل ﺗﺼﺤﯿﺤﺎت دﻗﯿﻖ اﺛﺮات اﺗﻤﺴﻔﺮ ﺑﺮ روي ﺗﺼﺎوﯾﺮ در ﻣﻨﺎﻃﻖ ﺑﺎ ﻣﻘﯿﺎس ﻣﺤﻠﯽ و ﺑﺰرگ ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﺗﻐﯿﯿﺮات ﺷﺮاﯾﻂ اﺗﻤﺴﻔﺮ و ﺧﺼﻮﺻﯿﺎت زﻣﯿﻦ ﻗﺎﺑﻞ ﭘﯿﺸﻨﻬﺎد اﺳﺖ.

Background and Objective: There was a possibility to study earth coverage on a large scale using remote sensing data. The support vector machines (SVM), artificial neural network (ANN) and maximum likelihood (ML) algorithms were used to Land cover classification on OLI sensors data and 4 kernels in Guilan province. Methods: Classifications were based on training samples of 10 different covers in the entire Guilan province. To improve the classification accuracy on OLI image data, the MODIS atmospheric products used in 6SV atmospheric correction model. The OLI atmospheric corrected image segmented to 219000 polygons based on homogeneity. In this study 2% of polygons were used to test and training samples by the random statistical method. Polygons labeled to classes by field survey. Findings: Applying ANN, SVM and ML algorithms on the OLI images after atmospheric corrected by 6SV model, the overall accuracy of classification improved 0.11%, 0.8%, and 1.9% respectively. The results indicated that the land cover map by RBF-SVM had overall accuracy and kappa coefficient with 75.6% and 0.72 respectively. In this algorithm accuracy of agriculture, range shrub land and water body classes were 93.16%, 72.55% and 96.57% respectively. The results of this study indicated that SVM algorithm improved overall accuracy 1.67% compared to the ML algorithm. Discussion and Conclusion: This research indicated that in land cover classification and mapping of Guilan province, the nonparametric SVM algorithm had more accurate than the ML parametric algorithm. According to the results of this research, it is suggested that atmospheric correction models should be used especially on the large and local images.