فصلنامه سنجش از دور و GIS ایران
سال سوم شماره 3 (پیاپی 11، پاییز 1390)

Due to expanding applications of the SAR (Synthetic Aperture Radar) images and the effect of Speckle noise in reducing the Interpretability of the SAR images, Speckle noise reduction is very important. In this paper, an Adaptive Filter has been proposed to reduce the speckle noise. In this filter, L1 norm (first moment) has been used to identify pixels that have been severely affected by Speckle phenomenon. In the Proposed method, the nearest degree of previous gray level --within the allowable limit-- instead mean of kernel has been placed in the pixels. In This paper it is tried to determine the best mode for the proposed filter as a damping factor in the used images. The results have been compared then to the results obtained by other widely-used adaptive filters including Lee, Gamma, Frost and Touzi. The results of this comparison showed that the proposed filter with the mean signal to noise ratio (SNR) above 14 offers better results than the above-mentioned filters with the maximum value 10.13.

Interpolation is among the most important areas in geosciences, because there is always a need to obtain reasonable large and accurate information from a finite set of data. Many methods of interpolation have been developed so far. Kriging is one of the most applicable and accurate among them. Also, Kriging has been classified into various sub-types; Such as: Ordinary Kriging, Indicator Kriging and Soft Indicator Kriging. Genetic Algorithm is a common evolutionary algorithm used to obtain an optimum result from a potential answer space. In this paper, Genetic Algorithm has been utilized to train some parameters of SIK relating to an objective function to result in the best answer. In this article, Ordinary Kriging, Indicator Kriging, Soft Indicator Kriging and Soft Indicator Kriging & Genetic Algorithm have been applied to a data set collected from 15 Synoptic stations in Mazandaran province, Iran. The data are the observations of monthly obtained precipitation of Bahman, 1387. By comparing the calculated and observed values in both Cross-Validation and Jackknife approaches in all of the above mentioned methods (4ones) it is proved that the combination of Soft Indicator Kriging and Genetic Algorithm is the most accurate approach. The mean of the absolute differences between the observed and calculated values of each method is the criterion by which the methods have been evaluated. This quantity for the Soft Indicator Kriging & Genetic Algorithm method is 10.3 mm and is the least one compared to the other methods (12.3, 13.3 and 12.8 mm respectively).

In the recent years a considerable attention has been carried out to explore the relationship between crashes and related factors such as land use, demographic characteristics and network parameters in urban areas. Trip production and attraction as the first-hand results of trip generation models along with total road kilometers might be proper surrogate predictors to estimate the crashes in aggregate spatial units. Generalized Linear Models are known as the common techniques to investigate the relationship between crashes and predictors based on estimating the global fixed coefficients. Geographically Weighted Poisson Regression is suggested to investigate the non-stationary effects in relationship between crash frequencies and contributing factors. The model was run using the trip production and attraction parameters and total road kilometers over 253 Traffic Analysis Zones in Mashhad and the existence of spatial autocorrelation was examined among residuals. The goodness-of-fit tests indicate a considerable improvement in the model performance comparing the global models. The results also show that the significant non-stationary state in relationship between crash frequency and selected predictors. Based on the results, there is a direct and positive relationship between the independent variables and dependent variable; although the coefficient magnitudes vary over different spaces. The examination of residuals for global and local models reveals that the spatial autocorrelation has successfully been removed from the local residuals.

To evaluate the impacts of polluted soils on human health, it is important to understand the of soil particle size distribution. Polluted soil particles are considered as pollutants and their transportation by water and wind depends on particle size. Applying the of new techniques such as spectrophotometry and remote sensing which enhance feasibility and swift determination of soils pollution risk could be useful due to high costs of soil sampling and laboratory measurements. This research has been carried out to determine spatial distribution of zinc concentration in soil particle size classes using IRS-LISS III reflectance in Southern Isfahan city. There fore, 100 compound soil surface samples have been collected randomly through the area. The samples have been air - dried and soil particle size classes 250-500, 125-250, 75-125, 50-75 and <50 μm have been determined, using appropriate sieves after dispersion the bulk soil samples by ultrasonic apparatus. Total Zn concentration has been measured using Atomic Absorption Spectrophotometer after wet digestion of samples in acid nitric. Results showed that concentration of Zn increases by decreasing soil particle size. Significant negative correlation coefficients have been found between Zn concentration in soil particle size classes and soil spectral reflectance in visible and near infrared bands of IRS-LISS III satellite. Stepwise multiple regression models have been used for estimating Zn concentration in soil particle classes through satellite data and spatial distribution of Zn concentration in soil particle size classes were mapped using stepwise multiple regression equations. Results showed possible estimation of Zn concentration to certain ranges of concentrations using multiple regression models and satellite data so that in size classes <2000, 50-75 and <50 μm concentrations lower than 400 mg kg-1, measured and estimated Zn concentrations are closed to 1:1 line. This is an indication of higher precision in estimating Zn-concentration at lower concentrations. The results also indicated increments of Zn concentration in all soil particle size classes by approaching to the mines.

Nowadays image acquisitions devices such as cameras are common between different peoples. Therefore most of peoples can record visual information at any time and place. There fore different ubiquitous services could be developed based on these devices. In this article a vision-based location system has been presented. In this proposed system, each user can access the map of the region by sending an image of a noticeable nearby building in the town. At the first step we need a database consisting of the images of noticeable buildings in the town. In addition to the building images, the location of the building must be saved in the database. Received images should be compared then with the images in the database, using a complex image processing algorithms. In this paper the Scale Invariant Feature Transform algorithm have been used. The most similar image to the received image is detected from the database by comparing the descriptors of key points extracted from image using the algorithm. The location of user is assumed to be the location of identified building in the database. At the last step the map of the region can be send to the user.

According to the importance of location sciences, in decreasing the costs and efficiency on presenting services, there are several researches in this context. Besides, Mathematical models have been presented to explain the general state of different location problems. Several objective functions relevant to each model, have been formulated. According to lack of the researches in context of dynamic locating and moving object allocation, this type of location/allocation problems are considered in this research. This type of problems is divided into three main categories. In this paper, one of the useful classes of the mentioned problems is formulated as an optimization problem and an efficient solution based on genetic algorithm is presented. Evaluating the proposed method, a sample problem is designed and solved using both proposed and traditional methods. The results of implementation have indicated the efficiency of the proposed method.

Nowadays, by developing the remotely-sensed systems and acquiring the hyperspectral data with the high spectral resolution, it is expected that phenomena have been accurately recognized, and as a result, clustering accuracy of corresponding land-cover classes have been increased. Despite this useful characteristic, spectral information with high dimension is a challenging problem in data processing and analyzing. That is why, by increasing the dimension of data, firstly clustering accuracy is improved to a threshold, and then it decreases. So in order to make an appropriate clustering, it is needed to choose the optimum bands of these images simultaneously or in other words, the two problems of clustering and choosing optimum bands should be considered together. In this paper a new method has been presented for clustering of hyperspectral data by PSO, in which both optimum centroids and the number of optimum bands of these images are achieved. High capability of PSO algorithm in global searching of the space problem, and at the same time, solving the multi-objective problems caused to use this algorithm for clustering the hyperspectral data. In this study, after surveying the concept of PSO-based clustering, the proposed method is presented based on solving clustering and choosing optimum bands simultaneously. Comparison of the obtained results for clustering of a scene of AVIRIS image has showed that clustering by PSO and proposed method with the accuracy of 74.22 and 76.57 respectively for Kappa coefficient gives better results than the other algorithms which have been used in this research.