علمی - پژوهشی
M Akhondi; M Mesgari; M. R Malek; O Askari Sichani
Volume 9, Issue 2 , December 2017, Pages 1-20
Abstract
Nowadays, heavy traffic is one of the major problems of living in big cities. In recent years, to overcome this problem, various solutions are proposed, many of which have been on the basis of general and comprehensive models. However, because of the essential complexity of urban environment and because ...
Read More
Nowadays, heavy traffic is one of the major problems of living in big cities. In recent years, to overcome this problem, various solutions are proposed, many of which have been on the basis of general and comprehensive models. However, because of the essential complexity of urban environment and because of the diversity of parameters affecting urban traffic, those models cannot represent the dynamic space of urban traffic, properly. In contrast to them, agent based approach is a promising approach for modeling of urban traffic. This is mainly because of its ability in modeling the interactions of traffic components, and in the modeling of the dynamic nature of urban environment. Much research has been made in the field of application of agent technology to the modeling of urban traffic. The majority of these researches are focused on a particular area of traffic phenomenon. Some of them are on providing traffic lights with some levels of intelligence. Others try to simulate the behavior and decision making of the drivers. In other cases, agent based modeling is used for simulation of dynamic vehicle routing systems using real-time traffic information. Nonetheless, less attention is paid to the more comprehensive modeling of traffic using intelligent agents. Therefore, in this research, an agent based model is proposed for improving the navigation of vehicles, on the basis of communicating traffic information amongst traffic components. The urban environment is modeled as a vector space. The model components include the two-way streets, intersections, traffic lights, origin and destination of cars. Environment comprises of intersections, streets between intersections, and streets between intersections and the origin/destination points. Active agents are the cars, traffic lights and traffic control center. In this agent based model, the green-red changing of traffic lights is controlled and programmed based on the traffic jam condition (number of cars) of the streets connected to the light. It is assumed that all vehicles are equipped with GPS and necessary communication media. The system is implemented using JADE platform and its class libraries. The data of a simulated traffic network is entered to the model. The main result of this study is a simple model of the basic part of the urban traffic, in which mobile vehicles and traffic lights have access to online traffic information. In this model, all three types of agents, i.e. cars, traffic lights and traffic control center, can communicate with each other. By defining some criteria, the impact of such communications and access to online information can are assessed. In other words, the results of different scenarios are evaluated using criteria such as traffic jam and average of traveling time. An important aspect of the model is that, although communicating with each other, all agents including drivers and traffic lights act and decide independently, i.e. without any centralized decision-making system. In this study, no GIS software is directly used. However, the behavior of vehicles and traffic lights are modeled on the basis of metric spatial relationships (distance calculations) and topological relations (connections of the street edges with each other and with traffic lights). In other words, in this study, a simple spatial environment and simple spatial behaviors are modeled. Spatial environment of two-way street and moving in them is represented by movements in a set of simple lines in the direction of X and Y axes. This model is the first step towards a more complete modeling of urban traffic. In this model, the spatial movements of vehicles are modeled as vectors. The lengths of these vectors are calculated using the assumed vehicle speeds, the distance between points, and simple estimations of traffic jams.
علمی - پژوهشی
S Salehi; M.J Valadan Zoej; M. Sahebi
Volume 9, Issue 2 , December 2017, Pages 21-40
Abstract
In applications related to environmental monitoring and disaster management, multichannel synthetic aperture radar (SAR) data present a great potential, owing both to their insensitivity to atmospheric and Sun-illumination conditions and to the improved discrimination capability they may provide as compared ...
Read More
In applications related to environmental monitoring and disaster management, multichannel synthetic aperture radar (SAR) data present a great potential, owing both to their insensitivity to atmospheric and Sun-illumination conditions and to the improved discrimination capability they may provide as compared with single-channel SAR. However, exploiting this potential requires accurate and automatic techniques to generate change maps from images acquired over the same geographic region in different polarizations or at different frequencies at different times. Furthermore, sensitivity to contextual information of each pixel reduces the error rates in labeling process, thus generates accurate change maps. The smoothing effect of despeckling and the isotropic formulation of the Markov Random Field model cause over-smoothing of the spatial boundaries between changed and unchanged areas in the final change maps. In order to reduce this drawback, edge-preserving MRF models could be integrated in the labeling process. This method improves the precision of edges at spatial boundaries and increases the change detection accuracy. In this paper, a contextual unsupervised change-detection technique (based on a data-fusion approach) is proposed for two-date multichannel SAR images. A Markov Random Field model is formulated by using “energy functions” that combines the information conveyed by each SAR channel, the spatial contextual information concerning the correlation among neighboring pixels and the edge information. In order to estimate the model parameters, the expectation–maximization algorithm is combined with the recently proposed “method of log-cumulants.” The proposed technique was experimentally validated with semisimulated data produced by ASAR-ENVISAT images. Experiments illustrate a significant improvement (average 12%) with the proposed technique over the other change detection approaches. Integrating edge information yielded accurate results in exploiting various levels of changes (low-medium-high) whereas contextual information and information conveyed by channels were unable to detect low and medium level changes. Considering the small number of iterations, computation time is reduced considerably. Generally the highest accuracy achieved by the proposed algorithm is 99/67%.
علمی - پژوهشی
F Aghighi; O.M Ebadati; H Aghighi
Volume 9, Issue 2 , December 2017, Pages 41-60
Abstract
Light Detection and Ranging (LiDAR) point cloud dataset and 3 dimensional (3-D) models have been extensively used for urban feature extraction, urban management, forestry management, managing urban green space, tourism management, robotics, and video and computer games' production. One of the main steps ...
Read More
Light Detection and Ranging (LiDAR) point cloud dataset and 3 dimensional (3-D) models have been extensively used for urban feature extraction, urban management, forestry management, managing urban green space, tourism management, robotics, and video and computer games' production. One of the main steps toward reaching accurate 3-D models is clustering and classification of LiDAR point clouds data. The main purpose of this research is to find out, particular machine learning techniques, which are promising for best learning and classification of LiDAR point cloud data in an urban area. Therefore, the performances of K-nearest neighbor (KNN), Decision Trees (D3), Artificial Neural Networks (ANN), Naive Bayes (NB), Support Vector Machine (SVM), and Markov Random Field (MRF) classifiers were evaluated on the LiDAR and aerial image dataset of Vaihingen, Germany, in the context of the "ISPRS Test Project on Urban Classification and 3D Building Reconstruction." In regard to the literature review, MRF model has not been used to classify LiDAR point cloud data in Iran. In this research, we utilized all the geometrical features, intensity values of LiDAR and aerial images as well as extracted eigenvalues based features to distinguish five urban object classes, including impervious surfaces, buildings, low vegetation, trees and cars. In order to compute eigenvalues using local point distribution, this paper introduces a new cubic structure, which has been not found in previous studies. The final results of 3D classification techniques in this research were 2D maps that evaluated by the benchmark ISPRS tests maps. The evaluation shows that the performance of MRF model with an overall accuracy of 88.08% and the kappa value of 0.83 is higher than other techniques to classify the employed LiDAR point clouds.
علمی - پژوهشی
V Ahmadi; A Alimohammadi; J Karami
Volume 9, Issue 2 , December 2017, Pages 61-78
Abstract
Management and planning of urban water supply in metropolis is very important. Development of the region urban and make cities to metropolis and increase of effective complex factor on water usage in the cities make consumption management and supply and distribution Water difficult. So rule extraction ...
Read More
Management and planning of urban water supply in metropolis is very important. Development of the region urban and make cities to metropolis and increase of effective complex factor on water usage in the cities make consumption management and supply and distribution Water difficult. So rule extraction plays an important role in exploring patterns over data and decreasing complex. Rough Set Algorithm, which was developed in 1980s by Pawlak, is a powerful and flexible method to deal with uncertain and ambiguous data which has been used in this research to extract dominant rules over data set. The method used in this paper is combination of the rough set and genetic algorithms from data mining methods to develop rule extraction and data classification of water usage in Tehran city as the studying area. Socio-economic, environmental, time and water consumption and management zones have been used as the explanatory variables for prediction of the water use that database divided to 2 part 60% for result extraction and 40% as test set. Independent test sets have been used for evaluation of the accuracy of the extracted rules. Results have shown that, combination of the genetic algorithms and Rough Set leads to extraction of more reliable rules. Classification accuracy of the extracted rules from Rough sets was 77 percent. But optimization of rules by combination of the genetic algorithm with Rough sets, resulted in classification accuracy of 88 percent in 6th generation with average speed of convergence. By using the same speed of convergence in the accuracy increased to 92 percent in 10th generation. According to the extracted rules, important effective factors on annual water consumption are respectively the resident population, water price, population density, family size, spatial location (latitude), education levels, and per capita green spaces.
علمی - پژوهشی
T Ensafi Moghaddam,; F. Khoshakhlagh; A.A Shamsipour; R Akhavan; T Safarrad; F Amiraslani
Volume 9, Issue 2 , December 2017, Pages 79-98
Abstract
Dust in the atmosphere and their interactions with precipitation have great impacts on regional climate where there are large arid and semiarid regions. Dust is one of the factors affecting precipitation. There are many ambiguities about the cause of the difference between amount of rainfall from an ...
Read More
Dust in the atmosphere and their interactions with precipitation have great impacts on regional climate where there are large arid and semiarid regions. Dust is one of the factors affecting precipitation. There are many ambiguities about the cause of the difference between amount of rainfall from an area to another area and from time to time. So that even with the spread of knowledge and technology yet still there is not completely specified the cause of these fluctuations. Nowadays, satellite images are broadly used for monitoring the effects of dust variations on the precipitation changes. Nowadays, satellite images are broadly used for monitoring the effects of dust variations on the precipitation changes. The aim of this study was to investigate the relationship between dust dynamic and precipitation variations. This research can be help to find the impact of dust occuarrances on precipitation changes in the South-West parts of Iran during thirty years by cluster analysis, remote sensing, and aridity zoning in GIS software. In this investigation, we analyzed data sets of daily average visibility(a proxy for surface aerosol concentration), daily No of reports frequency of dust occurance and daily precipitation at 45 meteorological stations during past 30 years(1986-2016)were obtained from the Iran Meteorological Organization. The consistent trends in observed changes in visibility, precipitation, and daily No of reports frequency of dust occurance appear to be a testimony to the effects of dust. In present study, we tried to determine the relationship between dust events data and measured precipitation changes in a ground stations. Therefore frequency of dust occurrence from 1986 to 2016 at 30 stations, compared with rainfall anomalies for South-West of Iran as a whole. Rainfall is expressed as a regionally averaged, standardized departure (departure from the long-term mean divided by the standard departure), but the axis of the rainfall graph is inverted to facilitate comparison with dust occurrence. Dust is represented by the number of days with dust haze. Then‚ dust days ratios which measured by number of days with dust in month and horizontal viewing which measured by number of days with visibility min < 2000 were compared by map of mean annual rainfall of stations in South-West of Iran. Precipitation maps were created using the inverse distance weighting interpolation (IDW). Too, In this study, (MODIS) Aerosols Optical Thickness(AOD)product is applied in order to estimate dust intensity. AOD images and MODIS/Terra Calibrated Radiances 5-Min L1B Swath 1km (MOD021KM) were utilized to assessment of move pattern of the dusts in the study area. Our results indicated that MODIS products could be a reliable tool to assess dust events patterns and to survey the concentration of particulate matter .So AOD images and MODIS/Terra Calibrated Radiances 5-Min L1B Swath 10km were utilized to assessment of special move pattern of the dusts frequency in the study area‚and indicated the opposite response of light rain to the increase in dust, have seen in mountainous and plain areas.
علمی - پژوهشی
I Yoosefdoo; A Khashei Siuki
Volume 9, Issue 2 , December 2017, Pages 99-116
Abstract
The use of groundwater plays an important rule for agricultural and drinking water purposes in the north of Iran especially in Koochesfehan region. In these areas, the excessive use of chemical fertilizers, especially nitrogen based ones, beside the inadequacy in the treatment and release of urban and ...
Read More
The use of groundwater plays an important rule for agricultural and drinking water purposes in the north of Iran especially in Koochesfehan region. In these areas, the excessive use of chemical fertilizers, especially nitrogen based ones, beside the inadequacy in the treatment and release of urban and industrial wastewater are some of the most effective parameter in groundwater pollution, especially about the concentration of nitrate. Therefore, identification and mapping of vulnerable aquifer areas, i.e. areas where pollutants can be penetrated and discharged from the ground surface to the groundwater system, is an appropriate management tool for preventing the pollution of groundwater resources. In this study, with the purpose of identifying vulnerable aquifers and areas with high nitrate content as the main vulnerability areas, by using 7 variables the Drastic method and by using the Aller weighing criterion, vulnerability index of the region was estimated. Then, by comparing the vulnerability index and the amount of nitrate measured in the zoned area, the correlation between nitrate and Drastic vulnerability index was calculated. The results showed that the vulnerability of the Astaneh-Koushfahan plain aquifer is located in four areas: 56.16% of the plain has a low vulnerability, 51.29% has a low to moderate vulnerability, 28.46% has a moderate to high vulnerability, 67.1% is vulnerable. It is too much. The correlation between the Drastic (vulnerability) index and the concentration of nitrate was 80%, which confirmed that nitrate was the main cause of vulnerability in this the aquifer. So, finding a method for estimating the amount of nitrate in present and future in this area with high speed and precision was assumed as the goal of this study. The amoun of nitrate were estimated with four artificial intelligence methods: artificial neural network, fuzzy model, support vector model and fuzzy-neural network. For this purpose, the seven Drastic variables data assumed as input parameters and the measured nitrate content in 30 different wells of the area were zoned by use of GIS software and divided into two categories of training and experimentation and they give as output parameters to all data-driven models. The results showed that all used artificial intelligence models give a good estimation of the amount of nitrate, but the neural network model had the best results, so that there was a correlation of 98% between computational nitrate and observation nitrate value. Finally, by choosing the model of the neural network as the superior model, it was tried to estimate the nitrate by decreasing the input parameters. The results showed that with 5 parameters of soil environment-unsaturated medium-saturated environment -water-hydraulic and eliminating two parameters of nutrition and topography, the correlation of estimated nitrate with the actual amount of measured nitrate is 0.90.
علمی - پژوهشی
A Shamsoddini; Sh Esmaeili
Volume 9, Issue 2 , December 2017, Pages 117-132
Abstract
Differentiating agricultural areas which are not covered by vegetation from bare lands as well as identifying bare lands from urban areas in medium spatial resolution images, e.g. Landsat imagery, are usually difficult and erroneous tasks which lead to the inaccurate classification results. Therefore, ...
Read More
Differentiating agricultural areas which are not covered by vegetation from bare lands as well as identifying bare lands from urban areas in medium spatial resolution images, e.g. Landsat imagery, are usually difficult and erroneous tasks which lead to the inaccurate classification results. Therefore, this study aims to present a new approach to increase the accuracy of the classification. For this purpose, different scenarios were applied based on different input attributes. The input attributes comprised of spectral bands, textural attributes, i.e. grey level co-occurrence matrix (GLCM), and two types of indices including spatial and thermal attributes proposed in this study. Three classification methods, maximum likelihood (ML), artificial neural networks (ANN), and support vector machine (SVM) embedded with different kernels, were applied to examine different scenarios. The results showed that SVM algorithm embedded with Radial basis function (RBF) results in better accuracy, with overall accuracy of 98.81% and Kappa coefficient of 98.25%, when all types of input attributes were combined together. Finally, the variable importance analysis by random forest feature selection indicated that the proposed indices played an important role to execute more efficient classification by SVM.
