ارزيابي توان تفكيكي روش‌هاي طبقه‌بندي پيكسل پايه داده‌هاي لندست 8 در تشخيص نوع پوشش اراضي مناطق كوهستاني، مطالعه موردي: حوزه آبخيز بشار

Evaluation of the resolution of pixel-based classification methods of Landsat 8 data for determining the type of land cover in mountainous areas, case study: Beshar Watershed

ﻫﺪف از ﭘﮋوﻫﺶ ﺣﺎﺿﺮ، ﺑﺮرﺳﯽ ﺗﻮان اﻟﮕﻮرﯾﺘﻢﻫﺎي ﻣﺨﺘﻠﻒ ﻃﺒﻘﻪﺑﻨﺪي ﻧﻈﺎرت ﺷﺪه و ﻧﻈﺎرت ﻧﺸﺪه دادهﻫﺎي ﺳﻨﺠﺶ از دور در ﺗﺸﺨﯿﺺ و ﺗﻔﮑﯿﮏ ﭘﻮﺷﺶ اراﺿﯽ ﺣﻮﺿﻪ ﮐﻮﻫﺴﺘﺎﻧﯽ رودﺧﺎﻧﻪ ﺑﺸﺎر ﺑﺎ اﺳﺘﻔﺎده از دادهﻫﺎي ﻟﻨﺪﺳﺖ 8 ﺑﻮده اﺳﺖ. ﺑﺪﯾﻦ ﻣﻨﻈﻮر، ﭘﺲ از ﺑﺮرﺳﯽ دﻗﺖ ﻫﻨﺪﺳﯽ و اﻧﺠﺎم ﺗﺼﺤﯿﺤﺎت رادﯾﻮﻣﺘﺮﯾﮏ و اﺗﻤﺴﻔﺮﯾﮏ دادهﻫﺎي ﻣﺎﻫﻮارهاي، ﻣﺠﻤﻮﻋﻪ داده ﺣﺎﺻﻞ از ﺗﺮﮐﯿﺐ ﺑﺎﻧﺪﻫﺎي اﻧﻌﮑﺎﺳﯽ )ﺑﺎﻧﺪﻫﺎي 7 ،6 ،5 ،4 ،3 ،2 و 8( و ﺣﺮارﺗﯽ )ﺑﺎﻧﺪ 10( اﯾﺠﺎد ﺷﺪ. ﺳﭙﺲ، ﻃﺒﻘﻪ-ﺑﻨﺪي ﭘﯿﮑﺴﻞ ﭘﺎﯾﻪ ﺑﺎ اﺳﺘﻔﺎده از اﻟﮕﻮرﯾﺘﻢﻫﺎي ﻧﻈﺎرت ﺷﺪه اﺣﺘﻤﺎل ﺣﺪاﮐﺜﺮ، ﻣﺎﺷﯿﻦ ﺑﺮدار ﭘﺸﺘﯿﺒﺎن، ﻓﺎﺻﻠﻪ ﻣﺎﻫﺎﻧﺎﻟﻮﯾﯽ، ﺣﺪاﻗﻞ ﻓﺎﺻﻠﻪ، ﺷﺒﮑﻪ ﻋﺼﺒﯽ، ﭘﺎراﻟﻮﺋﯿﺪ، ﻧﻘﺸﻪﺑﺮدار زاوﯾﻪ ﻃﯿﻔﯽ، واﮔﺮاﯾﯽ اﻃﻼﻋﺎت ﻃﯿﻔﯽ، ﮐﺪﮔﺬاري ﺑﺎﯾﻨﺮي و اﻟﮕﻮرﯾﺘﻢﻫﺎي ﻧﻈﺎرت ﻧﺸﺪه K-Means و IsoData اﻧﺠﺎم ﺷﺪ. دﻗﺖ اﻟﮕﻮرﯾﺘﻢﻫﺎ در ﺷﻨﺎﺳﺎﯾﯽ ﻫﺮ ﮐﺪام از ﮐﺎرﺑﺮيﻫﺎ ﺑﺮ ﻣﺒﻨﺎي ﺗﺤﻠﯿﻞ ﻣﺎﺗﺮﯾﺲ ﺧﻄﺎ، ﺑﺎ اﺳﺘﻔﺎده از ﻣﻘﯿﺎسﻫﺎي دﻗﺖ ﺗﻮﻟﯿﺪ ﮐﻨﻨﺪه، دﻗﺖ ﮐﺎرﺑﺮ و دﻗﺖ ﮐﻠﯽ ﺑﺮ اﺳﺎس ﻗﺎﻋﺪه ﺧﻄﺎي ﺣﺬف و اﺿﺎﻓﻪ و ﺿﺮﯾﺐ ﮐﺎﭘﺎ ارزﯾﺎﺑﯽ ﺷﺪ. ﻧﺘﺎﯾﺞ ﻣﺒﺘﻨﯽ ﺑﺮ ﻣﺎﺗﺮﯾﺲ ﺧﻄﺎ ﻧﺸﺎن داد ﮐﻪ ﻣﻨﺎﺳﺐﺗﺮﯾﻦ اﻟﮕﻮرﯾﺘﻢ ﺑﺮاي ﺗﻔﮑﯿﮏ و ﺷﻨﺎﺳﺎﯾﯽ ﮐﺎرﺑﺮي/ﭘﻮﺷﺶ زراﻋﺖ، ﺳﺎﺧﺖ و ﺳﺎز، ﺻﺨﺮه، ﺟﻨﮕﻞ، ﺑﺎغ، ﻣﺮﺗﻊ، ﭘﯿﮑﺮه آﺑﯽ و رﻫﺎ ﺷﺪه ﺑﻪ ﺗﺮﺗﯿﺐ، اﺣﺘﻤﺎل ﺣﺪاﮐﺜﺮ، ﻓﺎﺻﻠﻪ ﻣﺎﻫﺎﻻﻧﻮﯾﯽ، اﺣﺘﻤﺎل ﺣﺪاﮐﺜﺮ، ﻓﺎﺻﻠﻪ ﻣﺎﻫﺎﻻﻧﻮﯾﯽ، ﺷﺒﮑﻪ ﻋﺼﺒﯽ، ﻣﺎﺷﯿﻦ ﺑﺮدار ﭘﺸﺘﯿﺒﺎن، ﻣﺎﺷﯿﻦ ﺑﺮدار ﭘﺸﺘﯿﺒﺎن، اﺣﺘﻤﺎل ﺣﺪاﮐﺜﺮ اﺳﺖ. درﺻﺪ ﺻﺤﺖ ﮐﻠﯽ و ﺿﺮﯾﺐ ﮐﺎﭘﺎي اﻟﮕﻮرﯾﺘﻢﻫﺎ ﻧﯿﺰ ﻧﺸﺎن ﻣﯽدﻫﺪ ﮐﻪ ﭼﻬﺎر اﻟﮕﻮرﯾﺘﻢ اﺣﺘﻤﺎل ﺣﺪاﮐﺜﺮ، ﻣﺎﺷﯿﻦ ﺑﺮدار ﭘﺸﺘﯿﺒﺎن، ﻓﺎﺻﻠﻪ ﻣﺎﻫﺎﻻﻧﻮﯾﯽ و ﺷﺒﮑﻪ ﻋﺼﺒﯽ ﺑﺎ دﻗﺖ ﮐﻞ ﺑﻪ ﺗﺮﺗﯿﺐ 69/59 ،75/9 ،77/25 و 68/02 درﺻﺪ و ﺿﺮﯾﺐ ﮐﺎﭘﺎي ﺑﻪ ﺗﺮﺗﯿﺐ 0/63 ،0/69 ،0/72 و 0/58 ﻧﺴﺒﺖ ﺑﻪ ﺳﺎﯾﺮ اﻟﮕﻮرﯾﺘﻢﻫﺎ ﻋﻤﻠﮑﺮد ﺑﻬﺘﺮي از ﺧﻮد ﻧﺸﺎن دادهاﻧﺪ. ﺑﻪ ﻃﻮر ﮐﻠﯽ، ﻣﯽﺗﻮان ﺑﺎ اﻧﺘﺨﺎب و اﺳﺘﻔﺎده از ﻣﻨﺎﺳﺐﺗﺮﯾﻦ اﻟﮕﻮرﯾﺘﻢ ﻃﺒﻘﻪﺑﻨﺪي ﺑﺮاي ﻫﺮ ﻧﻮع ﮐﺎرﺑﺮي/ﭘﻮﺷﺶ در ﻣﻨﺎﻃﻖ ﮐﻮﻫﺴﺘﺎﻧﯽ و ﺳﭙﺲ، ادﻏﺎم ﻧﻘﺸﻪﻫﺎي ﻣﻨﻔﺮد ﮐﺎرﺑﺮي اراﺿﯽ ﺑﺎ ﯾﮑﺪﯾﮕﺮ، دﻗﺖ ﻃﺒﻘﻪﺑﻨﺪي را ﺑﺎﻻ ﺑﺮده و ﻧﺘﺎﯾﺞ ﺑﻬﺘﺮي ﻧﯿﺰ ﺣﺎﺻﻞ ﺷﻮد.

The aim of this study was to investigate the ability of different supervised and unsupervised classification algorithms of remote sensing data for detecting and separating of land cover on Beshar River Basin using Landsat 8 data. For this purpose, after checking the geometric accuracy and radiometric-atmospheric corrections on satellite data, the data set was created to the combination of spectral bands (bands 2, 3, 4, 5, 6, 7 and 8) and thermal (band 10). Next, pixel-based classification using supervised algorithms including maximum likelihood, support vector machine, mahalanobis distance, minimum distance, neural network, parallelepiped, spectral angle mapping, spectral information divergence, binary coding, and unsupervised algorithms including K-Means and IsoData was done. The accuracy of the algorithms for identifying each land use /land cover based on the error matrix analysis was evaluated using the producer's accuracy, user accuracy and overall accuracy based on the omission and commission errors, and the kappa coefficient. The results showed that the most appropriate algorithm for separation and identification of land use/land cover including agriculture, construction, cliff, forest, orchard, rangeland, water body and fallow is maximum likelihood, mahalanobis distance, maximum likelihood, mahalanobis distance, neural network, support vector machine, support vector machine, and maximum likelihood, respectively. The percentage of overall accuracy and Kappa coefficient shows that the four algorithms including maximum likelihood, support vector machine, mahalanobis distance and neural network with overall accuracy 77.25, 75.9, 69.59, 68.26 and the Kappa coefficient 0.72, 0.69, 0.63, 0.58, respectively, is better than other algorithms. Generally, the integration of appropriate classification algorithms in mountainous areas increases classification accuracy and will have better results.