فصلنامه سنجش از دور و GIS ایران
سال پانزدهم شماره 3 (پیاپی 59، پاییز 1402)

The effect of urban thermal islands due to intersections with major environmental challenges of the 21st century is one of the most important studies on environmental phenomena, and in this regard, the study of the land surface temperature gives a clear perspective of the thermal islands in cities, which, according to the warm and dry climate of Yazd, examines the status and factors affecting the land surface temperature in this city seem to be necessary. This research, using the spectrally and spatially fused image of Landsat-8, for August 2020, and using machine learning algorithms, tries to model the changes in land surface temperature by calculating different parameters related to urban land perspective. Based on the results of this study, the spectral-spatial fusion of Landsat-8 with Sentinel-2 by Pan sharpening, increased 10.7% of the overall accuracy and 16.5% of the Kappa coefficient in the classification of this image. The study also showed that most neighboring parameters associated with land cover are ranked 1 to 11 of influencing the land surface temperature of Yazd city. In this area, the proximity to bare lands in the radius of 100, 50, and 150 meters ranked 1 to 3 of the most important parameters affecting the land surface temperature respectively. This study showed that the change in land cover arrangement could affect the land surface temperature and changing the bare lands to the built-up areas, up to 1.1°C, to vegetation, up to 2.1°C, and changing 30% of bare land to vegetation, up to 1.6°C can reduce the average land surface temperature in Yazd. Also, this study showed that two different models of vegetation simulation in Yazd city showed that the "land-sparing " model could reduce the average land surface temperature in Yazd by 1.3° and the "land-sharing" model by 1.4°C.

In the early 1990s, radar interferometry was introduced and used as a useful tool in the study of all phenomena that cause land surface deformations. If the land surface deforms between two radar images, a surface displacement map can be created with millimeter resolution and accuracy. This paper reports the findings of the Sentinel1 –A data time series results using the SBAS algorithm to detect surface deformation in the Sangan iron ore mine. Sangan Iron Ore Mine is the largest open pit iron ore deposit in the Middle East. Due to mining activities, this mine has undergone many changes in terms of topography and geomorphology, which can intensify geomorphological processes. To detect and obtain the amount of land deformation, 48 SAR images of Sangan iron ore mine obtained by the European Space Agency's Sentinel 1-A satellite were used. The time series (2014-2020) obtained from the deformation in the range of placer mines were analyzed. The results show the average displacement rate of -20 to -35 mm per year and the maximum cumulative rate of deformations of -120 mm. Investigation of the cross-section in the two parts of the apex and the center of the alluvial fan in the placer mines during the period 2014-2017 shows the topographic changes well. To evaluating the reliability of the results, the results derived from SBAS have been compared due to the lack of data in the range of placer mines with the values measured by the total station related to the mountain unit in the years 2020-2014. The results showed that the rate of deformations from radar data using the SBAS algorithm compared to the leveling data has followed a similar pattern. However, there may be some error due to the different nature, ie in the leveling of elevation deformations measured for a point, but in interferometry the average rate is obtained from adjacent points.

With the development of remote sensing science, the use of hyperspectral images is becoming more widespread. Classification is one of the most popular topics in hyperspectral remote sensing. In the last two decades, a number of methods have been proposed to address the problem of hyperspectral data classification.In the present study, a structure based on learning capsule networks has been used to classify hyperspectral images, so that the network structure can have the most optimal generation of features by using a convolution layer and a capsule layer, and at the same time Avoid overfitting the on training data. The obtained results show the high quality of production features in the proposed structure. With the development of remote sensing science, the use of hyperspectral images is becoming more widespread. Classification is one of the most popular topics in hyperspectral remote sensing. In the last two decades, a number of methods have been proposed to address the problem of hyperspectral data classification.In the present study, a structure based on learning capsule networks has been used to classify hyperspectral images, so that the network structure can have the most optimal generation of features by using a convolution layer and a capsule layer, and at the same time Avoid overfitting the on training data. The obtained results show the high quality of production features in the proposed structure.In order to improve the classification accuracy, the feature extraction approach through the designed network and the classification by the Extreme Gradient Boosting was compared with the classification method by the global deep network. The proposed capsule approach consists of 3 basic layers: 1) Prime caps, which are capsules of size 8 and 32 with 9 × 9 filters and movement step 2, 2) Digitcaps with 10 16-dimensional capsules, and 3) fully connected layer. The results of examining two approaches for deep networking as well as combining capsule networks with XGBoost reinforcement tree algorithm were compared. Approaches such as SVM, RF-200, LSTM, GRU and GRU-Pretanh were considered to compare the proposed approach based on the configurations mentioned in their research.Up in addition to the study and quality measurement of production vector deep features by the proposed method in different classifiers, the ability of deep global networks in the application of classification should also be examined. The results of examining two approaches for deep network and also combining CapsNet with XGBoost show that by using the proposed combined method, images are classified with 99% accuracy on training data and 97.5% accuracy on test data.Up in addition to the study and quality measurement of production vector deep features by the proposed method in different classifiers, the ability of deep global networks in the application of classification should also be examined.The results of examining two approaches for deep network and also combining CapsNet with XGBoost show that by using the proposed combined method, images are classified with 99% accuracy on training data and 97.5% accuracy on test data.

The dramatic increase in construction in recent decades has been accompanied by an increase in the number of construction violations in urbanized areas and has overshadowed the urban management and planning system, so preventing unauthorized urban construction is one of the main problems of city managers. The current method of controlling construction violations includes field inspections based on human knowledge, which in addition to the need to spend exorbitant financial, time and human resources, may lead to collusion between builders and municipal inspectors or even failure to identify construction violations in a timely manner. In this regard, providing an intelligent and accurate method for identifying construction violations and targeting the search for construction patrols is more than necessary. The aim of this study is to provide an intelligent strategic model in monitoring violations. The present research is applied in terms of purpose and descriptive and causal in terms of method and the data has been collected by library and field methods. The results of this study indicate that by using the intelligent monitoring system, it is possible to intelligently monitor illegal constructs by processing the required image and data techniques, with the least presence of human agents and in a shorter time. The overall accuracy of 94% and the kappa coefficient of 71% for image classification in this system confirm the accuracy of the above results. It shows that in this method, the speed and accuracy of image classification, identification of changing buildings and identification of illegal constructions are much higher than physical and existing methods.

Temperature is the most important parameter for studying spatiotemporal phenological changes in plants. Thus, the current study was aimed to investigate the potential of MODIS land surface temperature (LST) data for mapping growing degree days (GDD) and different phenological stages of Bromus tomentllus and Astragalus effusus in Chaharmahal and Bakhtiari Province. MODIS extracted maps of maximum, minimum and mean temperature, GDD index and phenological stages from 2018 to 2019 during growing season were assessed against weather station data and also field-based phenilogical data using Pearson analysis in three regions with different altitudes. Results showed that MODIS LST and GDD maps had more than 91 and 99% correlations with field-based air temperature and GDD data, respectively (p<0.001). In early growing season, GDD values were less than 16 degree-days and they were more than 5200 degree-days in the late growing season which explained one and all the phenological stages of the studied species in the study area, respectively. The study findings indicated that MODIS data have high capability in spatiotemporal stratification of phenological stages of the Bromus tomentllus and Astragalus effuses plant species. The knowledge of different phenological stages is essential in species conservation and rangeland sustainable utilization, therefore, species phenology map can be used as an effective tool in rangeland management in the related organizations.

Forest fires worldwide cause severe damage to vegetation, soil and natural habitats, resulting in direct and indirect negative environmental impacts such as deforestation, climate change and drought. Therefore, identifying and determining the hazards of vegetation that suffer from fire is crucial for their management and development. The proliferation of remote sensing images such as the active fire products of the Terra and Aqua satellites over the past two decades has been one of the most essential methods in detecting these fires. However, the active fire product of the MODIS sensor in previous studies has shown that they alone do not provide good results in fire-affected areas. Therefore, it is necessary to evaluate vegetation with basic maps. The aim of this study was to investigate two types of plant products and to discover the active fire of MODIS sensor and FNF-JAXA forest and non-forest cover maps for better separation of burnt areas of vegetation in Iran between July 1 and 160 2020. The results show the highest area of fire on Julius 144 with more than 49 thousand hectares and Julius 128 with more than 45 thousand hectares. However, the largest area of the fire, forest cover is estimated at 120 to 160 in 2020 with more than 14 thousand hectares. Khuzestan province had the highest area of fires in the period under study that most of these areas in agricultural lands and the three provinces of Fars, Kohgiluyeh and Boyer-Ahmad and Bushehr had the highest area of fires in forest cover. The highest frequency of fires was observed in agricultural lands, the main reason for which could be human intervention. The evaluation of the results showed that the use of the FNF-JAXA product (accuracy of 87.4% and a Kappa coefficient of 0.85) compared to MODIS products (accuracy of 80.3% and a Kappa coefficient of 0.78) in the separation of forest areas has better capabilities. However, the ability of MODIS products to distinguish between pasture and agricultural vegetation is an important advantage, which the FNF-JAXA product does not have. In general, the findings of the research show that the MODIS product and FNF-JAXA maps can be used as reference maps to distinguish different types of vegetation that are subject to fire, in damage assessment and management.

Vegetation plays an important role in protecting water and soil resources, stabilizing carbon and improving air quality. In Middle Zagros, forest and pasture vegetation is very important in terms of protecting soil and water resources and sustaining economic activities. In this research, using the Google Earth Engine platform and Landsat 7 satellite images, the drought of Middle Zagros (Lorestan province) was monitored with vegetation indices NDVI, SAVI and VCI, as well as meteorological drought index SPI for the statistical period of 2020-2000. To calculate the SPI index, the precipitation data of 9 synoptic meteorological stations with appropriate spatial distribution and the length of the statistical period (2020-2000) were used, and the processing was done in DPI software. In order to calculate the plant indices, first, all the geometrically corrected satellite images of the ETM+ sensor of the Landsat satellite were called for Lorestan province for each year. At this stage, an average of 52 images were called for each year. Then the images with less than 5% cloud cover were selected and processed. The results of the VCI index showed that mainly the studied area was affected by mild drought during the statistical period of 2020-2000. The year 2008 had the highest amount of drought related to the middle class with 5880.6 hectares among the studied years. The results of the SPI index showed that there was a moderate drought in 2010, a severe drought in 2008 and 2017, a mild drought in 2006, and a severe drought in 2019. The results of NDVI and SAVI indices also show the increase of thin vegetation classes and areas without vegetation by 1.331679 and 115164 hectares, respectively, and the decrease of normal and dense vegetation by 446160.7 and 682.4 hectares respectively per year. 2008 was compared to 2006 and 2007. Based on the results of all three investigated indicators, the favorable conditions of vegetation cover and ecological threat were obtained in 2016, 2019 and 2020. The highest level of this coordination between SPI meteorological drought and vegetation indices was observed in 2008 and 2010 and to some extent in 2019. In general, the results show that the increase or decrease of vegetation can be caused by the occurrence or absence of drought, while other factors such as land use changes should also be considered.