Environmental Erosion Research
پژوهش هاي فرسايش محيطي
E.E.R.
Literature & Humanities
http://magazine.hormozgan.ac.ir
1
admin
2251-7812
2717-3968
10.52547/jeer
6561
8888
45855/11/3/90
fa
jalali
1397
2
1
gregorian
2018
5
1
8
1
online
1
fulltext
fa
استفاده از روشهای پیشرفته یادگیری ماشین در پایش فرسایش بادی در جنوب ایران
Advanced machine learning methods for wind erosion monitoring in southern Iran
روشهای نوین و دقیق در تهیّه نقشه فرسایش (سنجش از دور و سیستمهای اطلاعات جغرافیایی)
پژوهشي
Research
<strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">یکی از مهم­ترین فاکتورهای مؤثر در فرسایش بادی، تغییر کاربری/ پوشش اراضی است. پایش دقیق کاربری/پوشش اراضی و شواهد فرسایش بادی، در مناطق خشک و نیمه</span></span></strong><strong><span dir="LTR"><span style="font-size:12.0pt;">­</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">خشک اهمیت زیادی دارد. تفکیک پوشش­های اراضی حاصل از فرسایش بادی نظیر پهنه­های­ ماسه­ای و نبکاها، نیازمند استفاده از روش­های دقیق سنجش از دور است. در این تحقیق برای تهیه­ی نقشه­ی کاربری/پوشش اراضی در زمینه­ی فرسایش بادی، توانایی تکنیک­های پیشرفته­ی یادگیری ماشینی بر تصاویر لندست ارزیابی شد. بدین منظور، تصاویر لندست </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">7</span></span></strong> <strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">(2006)</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> و لندست </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">8</span></span></strong> <strong><span dir="LTR"><span style="font-size:11.0pt;">)</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;"> 2013</span></span></strong><strong><span dir="LTR"><span style="font-size:11.0pt;">(</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> از نظر هندسی و رادیومتریکی تصحیح شد. روش­های بارزسازی تصاویر، اعمال شد و با الگوریتم­های ماشین بردار پشتیبان با </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">چهار</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> نوع تابع کرنل خطی، چند­جمله­ای، تابع شعاعی مبنا و حلقوی و روش شبکه عصبی مصنوعی خودسازمان­دهنده­ی کوهنن، طبقه­بندی و با روش حداکثر شباهت مقایسه شد. با استفاده از آزمون­های جدایی­پذیری، بهترین ترکیب باند ورودی طبقه­بندی انتخاب شد. ارزیابی دقت نشان داد که بهترین نقشه با ترکیبی از باندهای خام و پردازش شده و با الگوریتم ماشین بردار </span></span></strong><strong><span dir="LTR"><span style="font-size:12.0pt;">RBF</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> (دقت کلی </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">%88</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> و </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">%87/90</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> برای تصاویر لندست </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">7</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> و </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">8</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">) حاصل می­شود. اختلاف دقت این روش با روش­های ماشین­­بردار خطی، چند جمله­ای، </span></span></strong><strong><span dir="LTR"><span style="font-size:12.0pt;">SOM</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">، حلقوی و </span></span></strong><strong><span dir="LTR"><span style="font-size:12.0pt;">ML</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> به ترتیب </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">5/1</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">، </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">9/2</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">، </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">3/8</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">، </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">4/12</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> و </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">4/16</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> درصد برای لندست </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">7</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> و به ترتیب </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">16/2</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">، </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">16/4</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">، </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">19/6</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;">، </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">89/13</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> و </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">67/14</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> درصد برای لندست </span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:11.0pt;">8</span></span></strong><strong><span style="font-family:b compset;"><span style="font-size:12.0pt;"> است. نتایج نشان داد که دقت طبقه­بندی با استفاده از ترکیب باندهای پردازش­شده و باندهای خام، در مقایسه با باندهای خام به تنهایی به­ میزان زیادی افزایش می­یابد. </span></span></strong>
<strong>Extended abstract</strong>
<ol>
<li value="NaN"><strong>Introduction</strong></li>
</ol>
Wind erosion is one the most important factors of land degradation in the arid and semi-arid areas and it is one the most serious environmental problems in the world. In Fars province, 17 cities are prone to wind erosion and are considered as critical zones of wind erosion. One of the most important factors in soil wind erosion is land use/cover change. Therefore, accurate mapping of land use/cover and wind erosion evidences in arid and semiarid regions is the utmost importance. Moreover, for discrimination of land covers resulting from wind erosion such as sand sheets and Nebka, we need accurate remote sensing methods. In this study, capability of the advanced machine learning techniques on Landsat 7 and 8 imageries in mapping land use/cover related to wind erosion is evaluated. <br>
<strong>2- Methodology</strong><br>
The study area is located in the Fars province, in the southern part of Iran, (from 28°07′15″ to 28°13′07″N and 52°07′36″ to 52°23′55″E, covering an area of 17,230 ha), which is considered as the most critical wind erosion area of the province. Landsat 7 (2006) and Landsat 8<span dir="RTL">)</span>2013) images were corrected radiometrically using Dark Object Subtraction method. Although images from USGS website are corrected geometrically, we checked the images using stream and road maps. According to the variations in land use/cover spectral behavior across the study area, it was difficult to define training samples representing thematic classes in a supervised classification procedure. Then different image enhancement techniques were applied. Classification stopped using Support Vector Machine with four different types of kernels including linear, polynomial, Radial Basis Function, sigmoid and Kohonen’s Self-Organizing Map neural network. Results were compared with Maximum Likelihood method. Using separability analysis, the best input band combination for classification was selected. The Overall Accuracy and Cohen’s Kappa coefficient, derived from the error of matrix which were used for the accuracy assessment of the final maps. <br>
<strong>3- Results </strong><br>
Results from accuracy assessment showed that the best map of the land use/cover in the relation to wind erosion was produced using a combination of original and processed bands and RBF vector machine (overall accuracy of 88 and 90.87 percent for L7 and L8, respectively). According to the separability metrics, the near infrared (NIR) and short infrared band (SWIR1), the WDVI, SAVI, LI indices, and processed bands by edge analysis in the aspect of E were finally selected as the best input band combination. The difference between accuracy of this method with linear, polynomial, SOM, sigmoid and ML methods were 1.5, 2.9, 8.3, 12.4, and 16.4 percent for L7 and 2.16, 4.16, 6.19, 13.89, and 14.67 percent for L8, respectively. In addition, results indicated that there was a significant change in wind erosion potential and land use/cover in relation to wind erosion in the study area in a short period of time. Rangelands were decreased by 73 percent and 10 percent of these areas are covered by sand sheets. More than half of rangelands were converted to agricultural lands. Insusceptible areas with surface crust or rocks were decreased by 59 and 2.39 percent, respectively.<br>
<strong>4- Discussion & </strong><strong>Conclusions</strong><br>
The accuracy of classification increased using a combination of processed and original bands in comparison with using original bands alone. This indicates the fact that processing image classification without paying attention to the quality of input bands, will not results in accurate classification map. One of the advantages of active learning algorithm is its less training samples requirement. This is very important for areas which are difficult to have access to them. Although there were not distinct and large sand dunes in the study area like what can be seen in desert areas of Iran, but discrimination of these small sand dune and nebkas were done accurately using the combination of original and processed bands of Landsat imageries and support vector machine methods. Goodarzimehr et al., (2012) also indicated that support vector machine was a better algorithm for discriminating lithology units comparing to maximum likelihood and neural network methods. Sandification was also recognized using remote sensing methods in this study which is one the indices of land degradation and wind erosion. Sand sheets showed and expansion mostly to the southeastern parts. The results indicated the change of rangelands into agricultural lands which will increase wind erosion potential. Low-efficiency irrigation systems combined with an increase in soil loss from arable lands leads to reduction in productivity. This is in line with findings by <a href="https://doi.org/10.1016%2Fj.proenv.2013.06.101">Minwer Alkharabsheh et al.</a> (<a href="https://doi.org/10.1016%2Fj.proenv.2013.06.101">2013</a>) who reported the progressive decrease of the agricultural areas and mixed rain-fed areas as the main reason of declining in soil erosion in Jordan. Generally, this study showed the capability of Landsat imageries and support vector machine learning in study of wind erosion potential in arid areas.
فرسایش بادی؛ کاربری/پوشش اراضی؛ لندست؛ ماشین بردار
Wind Erosion, Land use/cover, Landsat, Vector Machine
39
58
http://magazine.hormozgan.ac.ir/browse.php?a_code=A-10-575-1&slc_lang=fa&sid=1
Mahrooz
Rezaei
مهروز
رضائی
mahrooz.rezaei@shirazu.ac.ir
10031947532846005022
10031947532846005022
Yes
Shiraz University
دانشگاه شیراز
Abdolmajid
Sameni
عبدالمجید
ثامنی
sameni@shirazu.ac.ir
10031947532846005023
10031947532846005023
No
Shiraz University
دانشگاه شیراز
Seyed Rashid
Fallah Shamsi
سید رشید
فلاح شمسی
fallahsh@shirazu.ac.ir
10031947532846005024
10031947532846005024
No
Shiraz University
دانشگاه شیراز