جستجوی مقالات مرتبط با کلیدواژه « طبقه بندی نظارت شده » در نشریات گروه « آبخیزداری، بیابان، محیط زیست، مرتع »

Vegetation as a natural component plays a significant role in increasing permeability, improving soil, reducing evaporation, and reducing the runoff and thus reducing the possibility of flooding. The use of new technologies such as remote sensing and geographic information system to study plant ecosystems and prepare land cover maps is necessary to know the effectiveness of these tools and to identify the best methods of their use. The purpose of this research is to investigate the vegetation cover using the NDVI and compare the performance of three supervised classification methods, the maximum likelihood method, the minimum distance from the mean, and the parallelepiped method in a part of the Great Karun watershed. To this end, TM and ETM images of Landsat satellite were used in one interval and NDVI in a 10-year interval (May 2008 to May 2018) with the help of supervised classification and maximum likelihood algorithm. The above data were prepared and analyzed using ENVI4.2 software, and the effectiveness of each method was evaluated with the overall accuracy index and Kappa coefficient. Based on the results in the maximum likelihood method, the overall accuracy rate is 90.35% and the Kappa coefficient is 0.878, in the minimum distance method, the distance from the mean is 74.32% and its Kappa coefficient is 0.675, and in the parallelepiped method, the overall accuracy is 67.09% and the Kappa coefficient was calculated as 0.593. Based on the results, the maximum likelihood method has the highest level of accuracy in satellite data group classification. Moreover, the results showed that in the 10-year period in Dez, Karun, and Karkheh watersheds, the spectral reflectance related to vegetation has decreased by 7.4%, 10.64%, and 13.83%, respectively. The results of this research can be effective for the practical use of the analysis that was done in relation to the studies of runoff and flood. According to the process of vegetation changes due to natural or human factors, the need for proper management in this area seems necessary.

In the last decade, the use of remote sensing has played an important role in identifying and assessing natural disasters, especially floods. Among these techniques, the Support Vector Machine algorithm (SVM) and Change Detection technique can be mentioned. The main objective of this study was to evaluate the capability of these techniques in determining the effects of flood in Gori Belmak Wetland and Poldokhtar triple wetlands in the north of Molab and outlet of Poldokhtar watersheds in Lorestan Province, which was faced with flood in April 2019. The land use maps of the region were prepared by applying supervised classification method and the SVM on the Landsat 8 satellite image in the 2013, 2015, 2017 and 2019. Validation of the maps and techniques using indicators of kappa and overall accuracy, showed the high accuracy of maps prepared. The kappa coefficient was calculated to be 0.87, 0.84, 0.83 and 0.87 for the maps of the studied years and the overall accuracy was 90.02, 89.51, 88.11 and 90.32, respectively. By extracting the water class, the changes that occurred on the water body of the wetlands were detected. The results showed that Gori Belmak Wetland, undergo extensive changes due to reasons such as drought in 2015, increase of 112.08 ha of surrounding arable lands between 2013 and 2019, as well as topographic features, especially lower slope than the three wetlands. In 2019, with the storage of flood, this wetland increased to 47.08 ha compared to 2017 and reached an area of 146.15 ha. The similarity of the results obtained in this study with the results of the research conducted in the study area by the Copernicus Emergency Management Service (EMS) and the Geoinformatics Unit research team on the flood of 2019 indicates the high accuracy of the used techniques and results of the present research.

Identification and monitoring of surface water using remote sensing have become very important in recent decades due to its importance in human needs and political decisions. Therefore, surface water has been studied using remote sensing systems and Sentinel-1 and Sentinel-2 sensors in this study. In this paper, two data fusion approaches and decision fusion improve the accuracy of surface water identification for the two study areas in Mazandaran province, Iran, due to dense forests, turbid rivers, and land Agricultural products have been proposed. Data fusion approach by increasing the spatial resolution of Sentinel-2 optical sensor from 20 meters to 10 meters and using radar dataset and water indices in one layer, monitoring surface water with much higher accuracy by classification using supervised classifiers such as Support Vector Machine (SVM), Neural Network (NN), and Random Forest (RF) were performed. Then, by combining the results using the Maximum Voting (MV) method, the accuracy of the results compared to the data fusion approach increased by 4 percent to 5 percent. Since narrow and confused rivers are not easily identifiable and extractable, the proposed approach was able to extract an accurate map of surface water.

One of the important challenges and issues in land management programs is that of physical growth of cities in different directions over time. The transformation of non-urban watersheds into urban ones has led to some environmental challenges and problems. An overall management plan needs accurate spatial information at consecutive times. Remote sensing technology offers great help in identifying and detecting land use changes. The aim of this study was to detect land use changes using images collected from 1991 and 2001 Landsat 5 satellite¢s TM sensor and 2018 images provided by the Landsat 8 satellite¢s OLI sensor over a period of 27 years on Abshuran Watershed, the Kermanshah. Following the required procedures in the preprocessing stage, a change detection map was prepared with an acceptable accuracy with supervised classification in the ENVI5.3 software environment. The results indicated that during the 2001-2018 periods, about 1.17 % of the share of the rangeland and farmland decreased and that the urban and residential lands increased from 5.92 % to 17.19 % of the total area. Most of the land use change has occurred in the urban and residential areas, thus an increase in the impermeable areas causes an increase in natural disasters such as flooding in addition to some negative effects on natural resources.

Rapid population growth and the need to provide the necessities of life has caused human beings to change land use widely. In many cases these changes have been accompanied by disruption of balance in nature and providing the possibility of soil erosion. This study was carried out to investigate the effect of land use changes on increasing the intensity of wind erosion. Landsat satellite imagery and implementation of supervised classification by using SVMs were used to prepare land use map. Ground surface temperature was calculated, as one of the effective factors in wind formation, with single channel and split window algorithms in different land uses. In order to make connections between land use changes and wind erosion, effective climatic parameters such as temperature, relative humidity and maximum wind speed were investigated. The highest surface temperature was observed in poor and non-vegetation areas, and the lowest temperature was estimated in highlands and rangelands. Results showed an increase in average temperature as well as maximum and minimum absolute temperature and decrease in relative humidity. Over the past decade, the maximum wind speed in the study area showed a significant increase, and it has increased from eight m/s in 1990 to more than 20 m/s in the last decade. This shows that land use change by removing vegetation cover, as well as the uncontrolled increase of human constructions and creating thermal islands in the last 30 years, has a significant impact on the emergence of climate change and increasing wind speed.

Planning and optimal use of resources and controlling and unprincipled changes in the future, requires studying the extent of change and destruction of resources. In fact, planners for principled decisions must have a full knowledge of land use, detection, prediction of land use change and land cover in order to better manage natural resources in the long time. The aim of this study was to evaluate the accuracy of different supervised classification algorithms of basic and object-oriented pixels in land use extraction in Samalghan watershed in three periods of time 1987, 2002 and 2019. The results showed that the support vector machine algorithms for the images of 1987 and 2019 and the neural network for the 2002 image in the pixel-based classification method have the highest overall accuracy and kappa coefficient. Also, the most obvious change that can be seen by comparing the prepared user maps is the change in the level of land uses with the growth of residential areas, thus this expansion has been continuously associated with a decrease in rangeland land use. Thus, from the years of 1987 to 2019, the residential land use area increased by more than 9.197 km2 and dryland lands during these years increased by 130.89 km2, irrigated agricultural lands also increased from 44.45 km2 and Rangeland use has also decreased by 272.3 km2.

The construction of large dams such as the Alborz Reservoir Dam in spite of water supply in drinking and agricultural sectors is undoubtedly associated with large changes in natural ecosystems. Knowing and awareness of these changes is essential in order to plan and manage sustainable areas affected by the dam. simultaneous use of remote sensing and geographic information technology is one of the common methods for monitoring changes in the region. In this regard, the purpose of this study is to investigate the relationship between land cover / land use changes by constructing and dewatering of Alborz Dam using remote sensing and geographic information systems. To this end, two Landsat satellite images were used during the 22-year period (1994 and 2016) by applying radiometric and atmospheric correction as illustrating the changes in land cover / land use of the region before and after constructing. Then, , land cover/land use maps were produced by using the supervised classification method of the support vector machine and the accuracy of the classifications were evaluated and Analysis of time-varyings land cover / land use was also performed by producing of change detection map.In addition to, LQ index was used to study the spatial variations of different land cover / land use classes. According to the results, over two decades, the area of ​​forest cover increased by 22.7 hectares, the area of ​​agriculture and pasture, as well as construction areas decreased by 735.66 hectares and 17.82 hectares respectively and the area of water which has submerged the above classes, was 530.37 hectars.  Investigating the changes in the spatial pattern of land cover / land use showed that the most changes in the upstream side (dam reservoir) with LQ equal to 1.68, belongs to the water, which clearly indicates the effect of Alborz dam construction and dewatering on ​​land cover / land use changes at upstream side of the dam.

Monitoring the area and their marginal land uses changes in lakes are one of the strategies to achieve proper management of lakes and wetlands. In the present study to monitor Maharlu Lake area and its marginal land uses changes in Fars Province, the Landsat images of 1987, 2001 and 2013 were selected in wet and dry periods (March and September). By applying geometric and atmospheric corrections, the maximum probability algorithm was used for providing land use map. The Kappa coefficients for the September of 1987, 2001 and 2013 were 0.85, 0.88 and 0.89, and for April of 1987 and 2013 were 0.86 and 0.88, respectively. After providing land use map and validation, the method of comparing satellite images was used to monitor the changes. The changes were monitored by comparing Landsat images after classification. Results showed that the Maharlu Lake area has decreased by 82% in September 2013 compared to 1987. This reduction was 20% in March, 2013 compared to 1979. By reducing the water level of lake in September, bare soil and saline lands were increased by 39 and 37.5%, respectively. Agricultural and range lands have been decreased by 28 and 3.5% from 1987 to 2013, respectively. In general, the area under cultivation around the Maharlu Lake and the water level has been decreased significantly due to the lack of significant changes in rainfall during the long-term period and significant increase in evapotranspiration of Maharlu area during the last three decades. Due to recent droughts, bare soils and saline lands have been increased during this period. This has different consequences which require the proper management and planning of lakes and wetlands, especially during the drought periods.

In the present study, four supervised classification algorithms including Maximum Likelihood, Mahalanobis Distance, Minimum Distance and Neural Network with and without TIR1 were used to mapping land use of southern Khorasan province. Based on the results, the highest of overall accuracy and Kappa coefficient were calculated for the Maximum Likelihood algorithm with and without of TIR1. Using of TIR1 increased classification accuracy by Maximum Likelihood and Mahalanobis Distance algorithms; but using of TIR1 decreased classification accuracy by Minimum Distance and Neural Network algorithms, remarkably. Using of thermal data along with other spectral bands caused facilitation of discriminating classes with similar spectral characteristics. According to the land use map, bare land covered about 60% area of southern Khorasan province, generally more than 90% of the area of the province is involved by sparse land or weak vegetation cover which is prone to wind erosion.

Image classification is always one of the most important issues in remote sensing, and the obtained information from image classification is widely used in this field and other applications like urban planning, natural resource management, agriculture, etc. Since the main purpose of processing satellite images is preparing subjective and practical maps, choosing a suitable classification algorithm has an essential role. This paper studies the efficacy of Support Vector Machines (SVM) algorithm regarding satellite image classifications and compares it to artificial neural network algorithm. SVM is a group of classified and observed mechanical learning algorithms, used in remote sensing. In this study, SVM algorithms were employed for land use classification of Meymeh area using ETM landsat data. The classification via SVM was automatically performed by three types of linear Kernel, polynomial, and radial basis. Besides, the performance of this method was compared to that of artificial neural network classification method. Results showed that the average overall accuracy and Kappa coefficient of SVM algorithms, including linear Kernel, polynomial and radial basis, were respectively 9 percent and 12 percent more efficient than artificial neural network classification. Consequently, this study substantiates the efficiency and sufficiency of SVM algorithms in classification of remote sensing images.

Due to the human activities and natural phenomena, the face of the earth is constantly changing. The use of remote sensing technology is considered as the best tool for change detection and evaluation. With advances in this technology and satellite images with high spatial and spectral resolution as well as a variety of image processing techniques, the changes in land-use could be estimated and managed. In the present study, the land use changes of Kashan Salt Lake were investigated in a 12-year period. Landsat TM images of 1990 and Landsat ETM+ images of 2002 were used in this study and after carrying out the required actions in the preprocessing stage the map of change detection was produced by supervised classification. Results showed that 35.18% of the study area has changed and the highest percentage of the changes is related to the mulching land class. Although, several considerations, including spatial resolution changes and spectral power changes should be taken into account in interpretation and comparison, significant changes in the study area indicate the necessity of further studies to achieve more accurate results.

Land use classification using remotely sensed images is one of the most common applications in remote sensing, and many algorithms have been developed and applied for this purpose in the literature. This study investigates the efficiency of Support Vector Machines algorithms in image classification. Support Vector Machines (SVMs) are a group of supervised classification algorithms of machine learning that have been used in the remote sensing filed. The classification accuracy produced by SVMs may show variation depending on the choice of the kernel function. In this study, SVMs were used for land use classification of Ilam dam catchment using Land sat ETM+ data. The classification using SVM method was implemented automatically by using four kernel types, linear, polynomial, radial basis, sigmoid and the results were analyzed thoroughly. Results showed that SVMs, especially with use of radial, polynomial and linear function kernels, outperform the maximum likelihood classifier in terms of overall (about 10%) and kappa coefficient(about 15%) accuracies. So, this study verifies the efficiency and capability of SVMs in classification of remote sensed images.