نشریه علوم و فنون نقشه برداری
سال یازدهم شماره 4 (پیاپی 44، بهار 1401)

Panicle counts (PC) provide valuable information about yield prediction in sorghum but are expensive and time-consuming to acquire via traditional manual approaches. In this thesis, high-resolution RGB imagery acquired by UAVs has been used. The proposed method based on task-aware spatial disentanglement (TSD) has been modified to improve the performance of panicle detection. TSDPC has high accuracy in comparison to state-of-the-art techniques such as CenterNet and RepPoints.

Rapid urban growth, especially in developing countries, is causing a large number of urban planning problems. Although only three percent of the global land surface is covered by urban areas, approximately 54% of the world’s population lives in urban centers; according the latest estimates, by 2050 it will increase to nearly 65%. Accurate information on Urban Land Cover (ULC) types and their spatial distribution are of paramount importance for urban planning and management. To date, many studies have been conducted in the context of ULC mapping, and several methodologies and datasets have been used (e.g. land surveying and satellite data) in this regard. Under this background, generating ULC maps using land surveying method is considered as the most accurate technique, however, it is a costly and time-consuming task. Spending the least time and cost to produce these maps is one of the main challenges for city managers. To address this issue, the integration of satellite images and state-of-art classification methods has been received considerable attention in recent years. This study seeks to produce a 10 m resolution ULC map for Tabriz city, locating North East of Iran, using Sentinel-2 satellite data. The present study also aims to compare the potential of two advanced classifiers including Random Forest (RF) and Deep Neural Network (DNN) in ULC mapping. Five ULC classes including bare land, built-up areas, road, vegetation, and water were considered in this regard. As the number of trees (ntree) and the number of variables (mtry) are two main criteria applying the RF algorithm. In this study, ntree was set to 100 and the mtry was set to the square root of the total number of input features. In the case of DNN, a DNN model with six layers, including one input layer with 10 neurons (bands 2-8A and 11-13 of sentinel-2), four hidden layers with 200 neurons per layer, and one output layer (five ULC classes). In this study, the ReLU activation function was used for the hidden layers, softmax activation function was used for classifying information in the output layer. Our findings illustrated that the DNN algorithm by providing 95.2% overall accuracy outperformed RF (overall accuracy = 93.1%). Analyzing the performance of two algorithms regarding ULC classes showed that the DNN algorithm provided better results in bare land and built-up classes; the user’s accuracy and producer’s accuracy of bare land class were respectively 9.6% and 1% higher than those of RF. Regarding the built-up class, these metrics were also higher than RF (user’s accuracy = + 0.3% and producer’s accuracy = + 4.3%). In contrast, the RF algorithm performed better in extracting the road class; the user’s accuracy and producer’s accuracy of road class were 3.65% and 4.1% more than those of DNN, respectively. RF and DNN showed the same performances in classifying vegetation and water classes. In general, both algorithms provided good performances in ULC classification, however, the overall performance obtained by the DNN algorithm was substantially higher than RF. Because the performance of the DNN algorithm is better than the RF algorithm, we concluded that DNN is a valid alternative tool that should be considered for ULC mapping.

The selection of an optimal and appropriate place for temporary accommodation of people affected by natural disasters, such as flood, has long been of concern to Crisis Management Planners. Failure to properly locate these centers may result in heavy damage. The objective of the present study is to determine appropriate places for establishing temporary post-flood settlement centers in Mazandaran province. In order to achieve this objective, firstly, effective criteria for locating temporary accommodation were determined and the standard maps were prepared and normalized. In the present study, nine criteria for locating temporary accommodation centers have been applied. The Analytic Hierarchy Process (AHP) method has been applied to weigh the criteria by availing experts’ opinions and studying related articles. In the next step, in order to incorporate the criteria according to the weights calculated by the method of AHP, the method Weighted Linear Combination (WLC) was applied and the final map was categorized into four classes of “very appropriate”, “appropriate”, “not appropriate” and “very inappropriate”. In conclusion, the location of temporary post-flood settlement centers with respect to demographic points was considered in the two classes of “very appropriate” and “appropriate” from the previous stage using the P_Median function in the genetic algorithm. The algorithm searches for 15 locations among the candidate points for temporary accommodation. The centers are located in higher populated areas in two classes more desirable than other classes.

There are many instabilities in the earth's rotation due to celestial bodies, gravitational forces, and earth internal dynamics which make the calculation of Earth Orientation Parameters (EOP) and Polar Motion (PM) parameters a challenging task. In today's world due to the increasing requests for predicting EOP and PM parameters in a wide range of fields such as Astronomy, Geodesy, Oceanography, and Hydrography various methods are used. These days it is possible to calculate accurate values of EOP and PM parameters by means of global positioning system (GPS), very-long baseline interferometry (VLBI) and satellite laser ranging (SLR). The core reason for the short-term prediction of these parameters is the impossibility of calculating these parameters in real-time due to heavy preprocessing procedures. Hence, researchers are seeking to employ different methods for accurate short-term prediction of EOP and PM parameters. In recent years, non-parametric methods such as least square (LS) with autoregressive moving average (ARMA) and also singular spectrum analysis (SSA) have been used to estimate these parameters. Another method for the prediction of the aforementioned parameters was conventional artificial neural network (ANN). Currently, Deep Learning has become a popular field that attracts many researchers. Deep learning is a subset of artificial intelligence (AI) and machine learning that uses multiple layers and parameters in order to extract complex features from the inputs. It is widely used in computer vision and time series prediction applications. In this paper, we used three deep learning methods namely LSTM, CNN, and MLP in order to predict PM parameters (x and y parameters). Furthermore, we have used Least square harmonic estimation (LSHE) method in order to compare the final results with different networks. LSTM equipped with a short-term recursive memory. This recursive mechanism prepares LSTM for handling time series data. CNN extracts important features of input data by convolution multiplication and each convolution layer within CNN architecture produces feature maps as output. These Feature maps contain recognized patterns which will be used as input of next layer. CNN networks, which were primarily designed for the computer vision, are being used more and more in timeseries prediction applications. MLP networks are similar to conventional back propagation feedforward networks, however, new activation functions and optimizers could be used in MLP networks. For sufficient training of different architectures, we used 35 years of daily PM parameters from 1st January 1980 to 31 December 2015 and we predicted 40 days periods ahead for the future 5 years. For comparison of predicted values by different networks, we used mean absolute error (MAE) as a criterion and illustrate results in two tables. Also, we depicted different figures to show how networks are working. In addition, we used two LSTM, four CNN, and two MLP Networks with ADAM optimizer, ReLU activation function, and learning rate of 0.001- 0.0001 in order to select the best network with the lowest errors.  Moreover, the figures of predicted values vs actual values and plots of MAE for 40 days are shown on four figures for better comprehension of ultimate results. In the end, it turned out that LSTM networks outperformed CNN and MLP networks and in this network the final results are better than the others in most days. For the x parameter in the first, twentieth and fortieth days, the best MAE values are 0.41 mas, 5.58 mas, and 12.45 mas and the best values for RMSE are 0.49 mas, 6.69 mas, 15.05 mas, respectively. For the y parameter in the first, twentieth and fortieth days, the best values of MAE are 0.54 mas, 3.24 mas and 7.56 mas and the best values for RMSE are 0.68 mas, 4.72 mas, 9.22 mas, respectively. The final results show that the neural networks outperformed LSHE method and the accuracies of the deep learning networks are satisfying and LSTM and CNN networks are capable to predict values accurately.

Environmental pollution and disasters have gradually increased with population growth. The presence of oil resources in the seas and the incidents related to their discovery, extraction and transportation cause the formation of oil slicks on the sea surface, and the leakage of these petroleum products into the seas has irreparable environmental consequences. That is why monitoring the effects of these accidents is very important for public health. Satellite missions are a very effective tool for detecting pollutants such as oil spills. Artificial Aperture Radar Sensor (SAR) is an active microwave detection system that can be used to detect oil leaks with optical sensors installed on Landsat-8, Sentinel-2 and Ester satellite systems, taking into account cloud cover and satellite re-visit time at the same location. Be. In this study, the oil spill area due to pipe leakage in Khark Island was studied with Landsat-8 and Sentinel-1 satellite images. Various image processing techniques were applied to Landsat-8 bands to highlight oil spills in connection with the accident, such as morphology and convolution filters. We used Landsat-8 images to support the Sentinel-1 results. Oil spills were successfully detected by analyzing SAR data and Landsat-8 results, and by visually interpreting the results, the selected methods are consistent in terms of displaying oil spill areas.

In recent years, Road detection and road extraction from satellite images with the advancement and development of deep learning algorithms in the field of semantic segmentation has received more and more attention of researchers. In this regard, most of the studies have been done in the field of Road detection and road extraction using optical images and in these studies, few studies have been performed using radar images worldwide. Therefore, the aim of this study was to use a deeply Refined Deep Residual Convolutional Neural Network (RDRCNN) to evaluate and compare the accuracy of road extraction from Sentinel 1 radar images in Tehran and Shiraz metropolitan areas in equal conditions in terms of number of educational samples, validation and architecture. It is the same. In this study, to extract the road using DNN, the VV-VH color combination of Sentinel 1 radar images from 8 different cities (Tehran, Mashhad, Isfahan, Shiraz, Tabriz, Urmia, Baghdad and Beijing) was used. Finally, the RDRCNN model with a residual connected unit (RCU) and a dilated perception unit (DPU) was used for road training and extraction. The research findings indicate that the RDRCNN model has performed almost the same in the process of identifying and extracting roads in the two cities of Tehran and Shiraz, and in general, the above model has performed slightly better in the city of Shiraz. In terms of accuracy evaluation metrics, for Tehran images, the criteria were Recall 57.66%, accuracy 51.29%, F1 score 54.43% and overall accuracy 92.78%, and for Shiraz images Recall criteria 60.77%, accuracy 54.71%, F1 score 57.40% and overall accuracy of 95.63% were obtained. The findings of this study show the low accuracy of road training and extraction from Sentinel 1 radar images for two metropolitan areas of Iran. In general, by comparing the results of this study with previous studies, it can be seen that one of the most important reasons for the low accuracy of the results is the low width of roads in Iranian cities; However, due to the lack of necessary studies in the field of road extraction with Sentinel 1 radar images, it is not possible to comment definitively on the results and it is suggested that more studies be done in this field.

With increasing the number of large engineering structures in cities, experts are looking for a good solution for monitoring these structures to avoid great financial and human damages. From the past, leveling and ground surveying were carried out to measure the deformation of structures and ground displacements along the vertical direction; but these measurements are time-consuming and costly. Also, the using of precision instruments and deformation sensors are not suitable because of their high cost, time-consuming and complexity. Due to the ability of Radar images and Radar interferometry techniques in the field of monitoring the ground displacement, in this research, we are looking for evaluating the potential of this method for monitoring the dam deformation and displacements. To achieve this goal, we used two sets of radar data which are CosmoSkyMed-X and Sentinel-1A.   In the time series processing of these images, the PSI method was selected then the star graph and Deloney triangulation were used. In the next step, we used both linear and nonparametric models for displacement estimation. The results were evaluated by applying two different displacement models and finally, the model with higher temporal coherence was selected as the appropriate model and the other model was discarded. In processing the Mamlu dam images, the appropriate model for monitoring the displacements with S-1A Radar data was the nonparametric model and for CSK data was linear model. Due to the lack of ground data collection from Mamlu dam in the same period of time with radar data, to evaluate the accuracy and proof the obtained results, the results of two radar data sets (CSK and S-1A) were compared with each other and a very high agreement was observed between the results of these data sets that the amount of RMSE calculated for ASC data of these two sensors is equal to 0.7703 mm and for DSC data was calculated 0.9551 mm, which is a very high consistency of the results, which can be a reason for the accuracy of the extracted results.

The Differential Interferometric Synthetic Aperture Radar (DInSAR) can be considered an efficient and cost-effective method for monitoring ground subsidence because of its extensive spatial coverage and high precision. Because of orderly observations from a broad-range product, the new commissioning of the first Sentinel-1 satellite offers better support to operational scrutinies via DInSAR. In the present paper, the results of a Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) time-series analysis of 42 Interferometric Wide swath (IW) products of Urmia Lake Bridge in northwestern Iran acquired between November 2014 to April 2021 for both descending and ascending pass using the Sentinel-1A observation with Progressive Scans in azimuth (TOPS) imaging mode is studied. The SBAS processing was based upon the analysis of 111 small-baseline differential interferograms. The results demonstrate that the majority of regional ground subsidence rates in the research area ranged from 10 to 210 mm during the study period. Also, the maximum subsidence rate exceeded 210 mm/year. The Line of Sight (LOS) direction for descending pass is 91 mm/year for ascending the pass. The board view displays that ground subsidence is intense on the bridge. The largest subsidence center is located at the central points of the bridge. GPS data verified the SBAS-InSAR-derived result. The results include displacement in the horizontal and vertical directions for ascending and descending passes.

Spatial data infrastructure is a set of policies, standards, access networks, spatial data, organizations and people that facilitate and coordinate the various tasks of production, collection, storage, access and optimal use of spatial data in a specific area. In order to implement the provincial spatial data infrastructure, the provincial management and planning organization will be responsible for leading and coordinating between the executive organs of the province and the establishment of the provincial geoportal. In order to measure the success of spatial data infrastructure implementation in Ilam province, first the effective factors and indicators in the successful implementation of provincial SDI were identified and then weighed using the combined method of Dematel technique and network analysis process (DANP), which the indicators of structure, financial resources, specialization, education and culture, respectively were the most important. Then the implementation of each of the criteria in the province was examined. Finally, the success rate of spatial data infrastructure implementation in Ilam province was calculated. The results showed that the realization rate of SDI in Ilam province is 63%, which according to the existing conditions, the performance of this province can be at a good level. Finally, the weaknesses in the implementation of SDI in Ilam province identification and solutions to eliminate them and increase the success rate of SDI in this province were expressed.

In recent years, the population in Iran has been increasing. This growth in the population of large cities multiplies the demand for dwellings. The increase in demand, in turn, has led to the creation of high-rise buildings and numerous complex registration maps in these buildings. The current method for registering and storing various legal boundaries in Iran is based on two-dimensional maps. This method has its limitations for visualizing different registration boundaries that define legal spaces. For this reason, a new method to register the ownership of buildings accurately and efficiently is more than ever needed. With the advancement in the field of 3D modeling, especially Building Information Modeling, a lot of research is being done to record the three-dimensional legal boundaries using these models in different countries. In this article, for the first time, Building Information Modeling was used to record three-dimensional legal boundaries in the Iranian jurisdiction. To do this, first different types of registration boundaries in Iran were identified according to the regulations of the Real Estate Registration Organization of Iran and the building subdivision instructions in Iran. Then, to record these boundaries, Industry Foundation Classes (IFC) was developed and implemented on a prototype, which is a building in Tehran. In the evaluation stage, the registration boundaries in the two proposed methods and the current method were compared using a questionnaire. The results of this evaluation in different criteria showed the high capability of the proposed method in registering and visualization legal boundaries approved by the Real Estate Registration Organization so that more than 80% of the elite community participating in the questionnaire considered this method more appropriate than the current method.

The lack of efficient inventory tools is an old and well-known problem related to in situ forest measurements. Time of flight technology (TOF), using photogrammetric techniques and the ability to detect depth, allows the creation of 3D point clouds and the measurement of various objects, including tree stems. Therefore, this paper presents the accuracy of diameter measurement with TOF technology compared to calipers for six tree species: (Fagus orientalis Lipsky), (Quercus castanaefulia C.A.M. subsp), (Acer velutinum Boiss), (Carpinus betulus L.), (Alnus subcordata C.A.M.), (Parrotia persica C.A.Mey.), differing in stem shape and bark (20 of each species) were studied in the Darabkola Research Forest. Then, the diameters at a height of 1 and 1.30 metres of the tree trunks were measured with a vernier calliper and a point cloud was created using a Phab 2 Pro smartphone. The target diameters were measured using the CloudCompare software. The results showed that there was not much difference between the field recordings and the TOF measurement. However, Acer velutinum, Fagus orientalis and Quercus castanaefulia had fewer measurement errors than Parrotia persica, Alnus glutinosa and Carpinus betulus. Overall, the diameter RMSE at a height of 1 m and 1.30 m was 1.01 cm and 0.87 cm, respectively. According to the Bias index, the values measured with the TOF technology were higher than those measured with the caliper for most species were. Due to the results obtained and the many possibilities of TOF technology, such as the possibility to measure the diameter at different heights, the creation of RGB-3D, the high scanning speed and accuracy, low complexity, the brightness and the lower cost than other technologies, three-dimensional surveying is a good option for forest inventory verification.