فصلنامه سنجش از دور و GIS ایران
سال چهاردهم شماره 4 (پیاپی 56، زمستان 1401)

Autonomous landing is a key challenging in the domain of UAV navigation systems. Developing an autonomous landing system requires a precise estimation of the UAV pose relative to landing marker, particularly in vision systems this involves precise Helipad recognition. It seems that traditional approaches including cascade classifiers, image matching and segmentation techniques to have major challenges in different weather conditions and scales. On the other hand, convolutional neural networks (CNNs) have been introduced as a powerful tool in the visual recognition systems in the recent years but the high computational cost of this techniques, limited their performance in the low cost and light weight UAVs. The aim of this paper is to compare the convolutional neural networks and cascade classifier for helipad detection. The results show that CNNs are invariant under translation, rotation, scaling and occlusion. The detection accuracy of this method is 99.1 % which is 3 % more than cascade classifier while its running time is suitable for real time UAV applications.

Numerous studies on the phenomenon of fire over the past several decades have provided an extensive set of input data and implementation and evaluation methods. However, this vast array of results and research is structured to provide a roadmap to new users in the field and guidance on various applications and conditions that have not yet been analyzed. In other words, the absence of coherent research on the relative performance of different remote sensing processes in the fire is felt to produce various products or the resulting utilities. To fill this gap, a relatively comprehensive analysis of fire studies in remote sensing publications has been performed in this study. Some of the general factors evaluated in the pre, during, post-fire studies were the manipulation of input data, the review of algorithms, and their development, as these are factors that can be controlled by analysts to improve the Final accuracy of analyzes and results. One of the important issues in the field of fire after the identification and discovery of fire, due to the permanent changes in the structure and composition of vegetation, is to study how vegetation is restored and its growth rate during the years after the fire. According to a study of fire studies in the country, about 48% of them are related to the identification and spread of fire and the remaining 52% are related to resuscitation and recovery. In a review of research related to identification studies, it was found that approximately 5% of its share was done using learning methods and the remaining 43% was done using traditional methods. At the same time, of the study-related share of Resuscitation studies approximately 21% to examine vegetation and 31% of the soil under the fire surface. The findings of this study can be useful in helping researchers to make decisions in the selection of data and algorithms used according to the purpose of study, in different branches of studies associated with fire. However, in addition to these general guidelines, an analyst can consider personal preferences or the benefits of a particular algorithm that may be relevant to a particular program.

Due to the importance of meteorological data and limitations of data gathering from ground stations, remote sensing can play an important role in the preparation of these data. The purpose of this study was to quantitatively evaluate the Land Surface Temperature (LST) obtained from Moderate Resolution Imaging Spectroradiometer (MODIS) sensor images for estimating the maximum and minimum daily air temperature in the Taleghan watershed. For this purpose, the maximum and minimum daily air temperature data of three existing ground stations for the period 2009 to 2015 were obtained. Day and night LST and Normalized Difference Vegetation Index (NDVI) values ​​of MODIS were also prepared. The relationships between each of the effective variables and maximum and minimum daily air temperature in ground stations have been extracted using multiple linear regression method. The results showed that there was a significant correlation between maximum and minimum daily temperature of ground stations with day and night LST and NDVI from MODIS sensor. Therefore, these variables were used in regression relationships. The results of validation showed that the established relationships with all effective variables had the most accuracy. Therefore, the best model for estimating the maximum daily air temperature had , NSE and RMSE values ​​of 0.74, 0.74, and +4.7, respectively and for estimating the minimum daily air temperature had 0.71, 0.72 and +2.9, respectively. Therefore, by converting the surface temperature obtained from MODIS sensor images, the air temperature can be estimated with high accuracy on a daily and monthly scales for various studies.

Gathering accurate information for statistics requires high cost and precision. The time factor is also one of the important issues that should be seriously considered in statistics. Therefore, the use of sampling methods and satellite images will be a good alternative for this purpose. In the present study, the aim of the effect of different cluster sampling schemes in estimating the quantitative characteristics of the traditional forests of Olad Ghobad in Koohdasht township, Lorestan province using Sentinel 2 sensor images. To estimate the studied characteristics, 150 clusters in the form of six designs (triangular, square, star 1, linear, L-shaped, star 2) were implemented in the region. Then, in each subplot, the characteristics of the number and area of the tree canopy were measured. Afterimage preprocessing and appropriate image processing (principal component analysis, texture analysis, and different spectral ratios to create important plant indices), the corresponding digital values of the ground sample plots are extracted from the spectral bands and used as independent variables. Modeling was performed using nonparametric methods of random forest, support vector machine, and nearest neighbor. The results showed that the average density per hectare was 51 and the canopy area was 32.94%. The diagram of the mean squares of the error of the training and test data against the number of trees for the characteristic number per hectare and canopy showed that the optimal number of trees was obtained at approximately 75 and 350 points. The results of validation according to the percentage of squared mean squared error showed that for both density and canopy surface characteristics of random forest algorithm with linear and double star sampling designs with the squared percentage of mean squared error respectively (46.00%) and (10.44%) and Bias (-0.02%, 2.82%) along with cluster sampling designs linear and double star, respectively, had better performance in modeling. In general, the results showed that the use of different cluster sampling schemes, nonparametric modeling methods, and Sentinel2 sensor images can better performance estimate the quantitative characteristics of Zagros forests.

In recent years, with the change of use and development of agricultural lands in the country's basins, the rate of erosion and sediment production has increased. Given that in most sub-basins, the long term data of sedimentation stations have not been recorded, it is difficult to estimate the amount of sedimentation and erosion. The objectives of this study was to determine the factors influencing erosion and sedimentation and to determine the quantitative values of erosion in the Khorkhoreh watershed, Kurdistan, Iran. For this purpose, first, using topographic maps, geology and aerial photographs in GIS environment, type and shape maps of erosion were prepared and evaluated by field survey. Based on the MPSIAC model, the nine factors influencing erosion for all sub-basins were identified separately and the scores of each factor were determined. By summing the factors, the degree of sedimentation for each sub-basin was determined and the amount of sedimentation and special erosion and total erosion in each sub-basin were calculated. The results showed that the topographic factors and the current state of erosion have the most role and the weather factor has the least role in the sedimentation rate of the basin.Moreover, 92% of the total basin has a high degree of sedimentation in the fourth order erosion class. The amount of Sediment Delivery Ratio of the basin (SDR) varies between 32 and 50 percent. The lowest and highest specific erosion rates in different sub-basins were 10 and 35 ton/ha.yr, respectively. Also, the amount of special sediment and special erosion of the whole basin was 6.4 and 17.4 ton/ha.yr, respectively.

Drought monitoring is critical for early warning of drought hazard. This study  is  attempted to develop remote sensing drought monitoring index (VDI), based on Accuracy of different bands of Moderate Resolution Imaging Spectroradiometer data MODIS, VDI focuses about the vegetation water stress. Spectral studies have demonstrated that due to the large absorption by leaf water, shortwave infrared reflectance (SWIR) is negatively related to leaf water content. Being sensitive to leaf water content, SWIR is widly utilized to construct various remote-sensing indices for example VDI for reflecting vegetation water content, . In this study, Vegetation Drought Index (VDI) was evaluated Based on the sensitivity rate to moisture by shortwave infrared reflectance bands SWIR 5 and 6 (VDI5 and VDI6). The data included the MODIS sensor images from Terra satellite in a period of nineteen years from 2000 to 2018 and Precipitation data are obtained from the Global Land Data Assimilation System (GLDAS), in Sistan & Balouchestan Province, Pearson correlation coefficient was used to evaluate the accuracy of the Drought spatial distribution maps calculated based on the two bands.Results indicate high significant correlation rate between VDI6 and Precipitation data . Study also showed that shortwave infrared band 6 (SWIR) is more sensitive to agricultural drought than band 5,in Sistan and Baluchestan province . The study recommends  to use VDI index with and 6 for better early detection and monitoring of agricultural drought in operational drought management programmes.

Leaf chlorophyll and nitrogen, due to their important role in photosynthesis are among the major biological parameters of plant physiological status. The ability to quantify chlorophyll and nitrogen can provide important information for precision agricultural activities, plant and agricultural resource management planning, and modeling ecosystem services and production capabilities. This study aimed to assess the capability of indices for estimating the amount of chlorophyll and nitrogen in wheat using spectral data at the canopy level and also determine the most suitable spectral regions and absorption features for this purpose. This research was carried out in a greenhouse environment and the spectroscopic measurements were performed using ASD Fieldspec-3 full-range spectral spectroradiometer. Four plant band indices were classified into two groups of ratio- (NDVI, RVI, and DVI) and soil-based indices (SAVI2) for the raw spectrum and the first derivative of the spectrum for the total samples, and the results were compared. The parameters of position, depth, area, asymmetry and width were calculated for seven absorption features extracted from continuum-removed spectra, and the correlation of these indices with chlorophyll and nitrogen content of wheat was examined. The results showed that SAVI2 had a stronger correlation (RMSE = 0.12, R2 = 0.85) with the chlorophyll content NDVI (RMSE=0.30, R2=0.69) had a higher correlation with the nitrogen content, while using the first derivative with NDVI provided better results. Moreover, area and depth parameters of 430-760 nm absorption spectrum were the best indicators for estimating the amount of chlorophyll and nitrogen in wheat, respectively.