fa برآورد سطح زیر کشت گندم (Triticum aestivum L.) با استفاده از تصاویر ماهواره‌ای سنتینل-2 (مطالعه موردی: شهرستان شوشتر) تخصصي Special پژوهشي Scientific & Research <span style="font-size:11pt"><span style="text-justify:kashida"><span style="text-kashida:0%"><span style="line-height:19.0pt"><span style="direction:rtl"><span style="unicode-bidi:embed"><span calibri="" style="font-family:"><b><span lang="FA" style="font-size:10.0pt"><span b="" style="font-family:" zar=""><span style="letter-spacing:-.2pt">داشتن آمار و اطلاعات به‌روز از سطح زیر کشت محصولات کشاورزی یک ضرورت در طراحی الگوی کشت، مدیریت منابع آب و ارزیابی اثرات زیست‌محیطی ناشی از آن است و در این رابطه سنجش از دور یک ابزاری سودمند محسوب می‌شود. در این تحقیق سطح زیر کشت گندم در شهرستان شوشتر استان خوزستان با استفاده از تصاویر ماهواره&shy;ای سنتینل-2 برآورد شد. به‌منظور شناسایی مزارع گندم، سری زمانی شاخص تفاضل پوشش گیاهی نرمال شده منطبق با فنولوژی گیاه گندم در منطقه، محاسبه شد و مبنای تهیه نمونه&shy;های تعلیمی قرار گرفت. در ادامه از دو روش طبقه‌بندی حداکثر احتمال و ماشین‌بردار پشتیبان برای طبقه‌بندی تصاویر و تفکیک مزارع گندم استفاده شد. نتایج ارزیابی صحت نشان داد که روش ماشین‌بردار پشتیبان با دقت کلی 5/98 و ضریب کاپای 5/96 دقت بالاتری نسبت به روش حداکثر احتمال داشته و به‌عنوان مبنای محاسبه مساحت مزارع گندم قرار گرفت. نتایج نشان داد که مساحت اراضی زیر کشت گندم 48233 هکتار به‌دست آمد که تفاوت اندکی با برآوردهای موجود داشت. بر اساس نتایج این تحقیق می&shy;توان اظهار کرد که استفاده از روش سری زمانی شاخص تفاضل پوشش گیاهی نرمال شده در ترکیب با روش طبقه&rlm;بندی ماشین‌بردار پشتیبان، امکان محاسبه سریع و دقیق سطح زیر کشت گندم را فراهم کرده و می‌توان از این روش برای سایر محصولات کشاورزی و برنامه&shy;ریزی الگوی کشت در منطقه استفاده نمود.</span></span></span></b><b><span dir="LTR" style="font-size:10.0pt"><span bold="" new="" roman="" style="font-family:" times=""><span style="letter-spacing:-.2pt"></span></span></span></b></span></span></span></span></span></span></span><br> &nbsp; <span style="font-size:11pt"><span style="line-height:15.0pt"><span calibri="" style="font-family:"><b><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">Introduction:</span></span></b><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times=""> Up-to-date information on growing area is necessary for cropping pattern, and remote sensing is a useful tool in achieve this </span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">goal (</span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">Hunt <i>et al.,</i> 2019</span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">)</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">. Among </span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">various</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times=""> image classification algorithms and vegetation indices </span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">for </span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">crop type identification, machine learning methods and </span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">normalized difference vegetation index (</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">NDVI) have had better results (</span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">Mousavi <i>et al.,</i> 2020</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">). In addition, attention to plant phenological stages and using multi-temporal images have led to more accurate crop distinction. The aim of this study was estimation of wheat growing areas in Shushtar county in Khuzestan province in Iran</span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">.</span></span><span lang="EN" style="font-size:10.0pt"><span style="font-family:"Times New Roman",serif"></span></span></span></span></span><br> <span style="font-size:11pt"><span style="line-height:15.0pt"><span calibri="" style="font-family:"><b><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">Materials and </span></span></b><b><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">Methods</span></span></b><b><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">:</span></span></b> <span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">T</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">ime series of NDVI index was obtained from eight Sentinel-2 satellite images from November to June, which correspond to the wheat growth stages. Then, maximum likelihood and support vector machine algorithms were used to identify wheat fields. To implement the classification methods, training and test samples were needed, which were obtained by matching the NDVI diagram of the wheat crop growth stages and field surveys.</span></span><span dir="RTL" lang="AR-SA" style="font-size:10.0pt"><span style="font-family:"Times New Roman",serif"></span></span></span></span></span><br> <span style="font-size:11pt"><span style="line-height:15.0pt"><span calibri="" style="font-family:"><b><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">Results:</span></span></b><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times=""> Error matrix was used to evaluate the classification results. Based on this, the support vector machine with overall accuracy and Kappa coefficient of 98.5 and 96.5 percent, respectively, had higher accuracy than the maximum likelihood with overall accuracy and Kappa coefficient of 97.8 and 95 percent, respectively. In addition, considering the wheat crop phenological stages using the time series of NDVI in </span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">training</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times=""> samples, selection increased the classification</span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">s</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times=""> accuracy. Based on the support vector machine results, the total wheat growing areas was estimated to be 48233 hectares, which showed a small deviation from the available </span></span><span style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">surveys</span></span><span lang="EN" style="font-size:10.0pt"><span new="" roman="" style="font-family:" times="">.</span></span><span lang="EN" style="font-size:10.0pt"><span style="font-family:"Times New Roman",serif"></span></span></span></span></span><br> <span style="font-size:10pt"><span style="line-height:15.0pt"><span style="unicode-bidi:embed"><span calibri="" style="font-family:"><b><span lang="EN" new="" roman="" style="font-family:" times="">Conclusion:</span></b><span lang="EN" new="" roman="" style="font-family:" times=""> Using the pattern of NDVI time series to obtain training samples showed the efficiency of plant </span><span new="" roman="" style="font-family:" times="">indices</span><span lang="EN" new="" roman="" style="font-family:" times=""> in estimation of wheat growing areas according to the crop phenological stages. In addition, the results revealed that the support vector machine classification was more accurate than the maximum likelihood in the study areas, and it was considered as a base method. One reason for the appropriate performance of the support vector machine </span><span new="" roman="" style="font-family:" times="">was</span><span new="" roman="" style="font-family:" times=""> appropriate distribution and adeqaute number of </span><span new="" roman="" style="font-family:" times="">training</span><span lang="EN" new="" roman="" style="font-family:" times=""> samples based on wheat phenological stages. It can be concluded that the time series of the normalized difference vegetation index in combination with the support vector machine classification provided the possibility of quick and accurate estimation of the wheat growing areas in Shushtar county in Khuzestan in Iran</span><span new="" roman="" style="font-family:" times="">. </span><span lang="AR-SA" dir="RTL" style="font-family:"Times New Roman",serif"></span></span></span></span></span><br> <span style="font-size:10pt"><span style="line-height:15.0pt"><span style="unicode-bidi:embed"><span calibri="" style="font-family:"><b><span lang="AR-SA" dir="RTL" style="font-family:"Times New Roman",serif"></span></b></span></span></span></span><br> <br> &nbsp; حداکثر احتمال, سری‌های زمانی, شاخص NDVI, گندم و ماشین‌بردار پشتیبان Maximum likelihood, NDVI, Support vector machine, Training samples and Wheat 258 271 http://agrobreedjournal.ir/browse.php?a_code=A-10-1567-1&slc_lang=fa&sid=1 Fatemeh Modiri فاطمه مدیری modiri@asnrukh.ac.ir 100319475328460010888 100319475328460010888 No Agricultural Sciences and Natural Resources University of Khuzestan, Mollasani, Iran دانشگاه علوم کشاورزی و منابع طبیعی خوزستان، ملاثانی، ایران Mohammad Moazami محمد معظمی moazami@asnrukh.ac.ir 100319475328460010889 100319475328460010889 Yes Agricultural Sciences and Natural Resources University of Khuzestan, Mollasani, Iran دانشگاه علوم کشاورزی و منابع طبیعی خوزستان، ملاثانی، ایران Kamran Almasieh کامران الماسیه almasieh@asnrukh.ac.ir 100319475328460010890 100319475328460010890 No Agricultural Sciences and Natural Resources University of Khuzestan, Mollasani, Iran دانشگاه علوم کشاورزی و منابع طبیعی خوزستان، ملاثانی، ایران