A probabilistic SVM approach for hyperspectral image classification using spectral and texture features

New hyperspectral sensors can collect a large number of spectral bands, which provide a capability to distinguish various objects and materials on the earth. However, the accurate classification of these images is still a big challenge. Previous studies demonstrate the effectiveness of combination of spectral data and spatial information for better classification of hyperspectral images. In this article, this approach is followed to propose a novel three-step spectral–spatial method for classification of hyperspectral images. In the first step, Gabor filters are applied for texture feature extraction. In the second step, spectral and texture features are separately classified by a probabilistic Support Vector Machine (SVM) pixel-wise classifier to estimate per-pixel probability. Therefore, two probabilities are obtained for each pixel of the image. In the third step, the total probability is calculated by a linear combination of the previous probabilities on which a control parameter determines the efficacy of each one. As a result, one pixel is assigned to one class which has the highest total probability. This method is performed in multivariate analysis framework (MAF) on which one pixel is represented by a d-dimensional vector, d is the number of spectral or texture features, and in functional data analysis (FDA) on which one pixel is considered as a continuous function. The proposed method is evaluated with different training samples on two hyperspectral data. The combination parameter is experimentally obtained for each hyperspectral data set as well as for each training samples. This parameter adjusts the efficacy of the spectral versus texture information in various areas such as forest, agricultural or urban area to get the best classification accuracy. Experimental results show high performance of the proposed method for hyperspectral image classification. In addition, these results confirm that the proposed method achieves better results in FDA than in MAF. Comparison with some state-of-the-art spectral–spatial classification methods demonstrates that the proposed method can significantly improve classification accuracies.     

Exact and Approximate Solutions of Thermoelastic Stresses in Functionally Graded Cylinders

Thermoelasticity behavior of functionally graded thick hollow cylinders with power law and exponentially variations of material properties versus radius is studied analytically. Temperature, displacement and thermomechanical stress distributions are obtained and discussed. Approximate homogeneous multilayer semi-analytical method is used successfully with adequate accuracy and finite layers. Different combinations of ceramics and metals are checked out for thermoelastic stresses due to high temperature and internal pressure. We conclude that the approximate method is simple and has appropriate results. The effects of high temperature on the stresses are more important than the high internal pressure.     

Bars on blocks: A cellular automata model of crime and liquor licensed establishment density

Criminologists have extensively researched the problems generated by licensed establishments. Violent offending and disorderly behavior resulting from pubs, taverns, dance clubs and bars are of particular interest to this field of study. The relative density of these liquor establishments has been found to be associated with the level of violence and disorder in surrounding areas. A complex systems approach can be used to further understand the dynamic interplay between licensed establishments, violent offending and disorder, and urban planning decisions. The model presented here utilizes cellular automata as the mathematical framework to view the varying impact of liquor licensing density on crime. This study uses a sample of liquor establishments and crime data from the City of Vancouver in British Columbia. The cellular automata model incorporates transition rules which govern the change of city blocks from low-risk blocks to high-risk blocks. The results represented by a 50 × 50 cellular grid show that high-risk blocks multiply when liquor licenses are grouped. Two scenarios are presented to contrast the impact of grouping high-risk blocks which contain more liquor establishments and dispersing such blocks. A third scenario demonstrates how increasing the positive influence in a grouped scenario stops high-risk blocks from taking over the entire grid. Future iterations of this model will incorporate census data, public transportation data, land use data and entertainment districts from other cities to further analyze the effect of licensed establishments on the distribution of crime.     

Machine scheduling in underground mining: an application in the potash industry

In this paper, a scheduling problem that occurs in potash mining is introduced, where a block excavation sequence has to be found taking into account a limited number of underground machines as well as safety-related restrictions. The aim is to minimize the maximum completion time of excavations, i.e., the makespan. The resulting problem can be transformed into a hybrid flow shop scheduling problem with reentry, unrelated machines, and job-precedences. A mixed-integer linear model is presented and small-scale instances are solved with CPLEX. In order to tackle medium- and large-scale instances heuristically, a basic and an advanced multi-start algorithm are developed, based on a specific priority rule-based construction procedure. In addition, a modified version of the Giffler and Thompson procedure is applied. Computational experiments are conducted on problem instances derived from real-world data in order to evaluate the performances of the proposed solution procedures.     

Numerical modeling the effects of overloading and underloading in fatigue crack growth

In this paper, the effects of overloading and underloading on fatigue crack growth were investigated. Numerical modeling was done by using finite element software. In this software, without using the remeshing technique, effect of crack tip plasticity in fatigue crack growth life was analyzed. Plasticity effects were considered by using three methods: COD method, U correction factor and J-integral. Calculated results for crack growth rates were compared with experimental data in literature. Results were obtained by COD method and U correction factor have good agreement but results form J-integral have not. We also study the effects of stress ratio (R) in plane stress and plane strain conditions. With increasing R, the fatigue life was increased. The extent of crack retardation is greater under plane stress than plane strain conditions. Underloading has not significant effects on fatigue crack growth rate. Underloading cause a little variation in plastic zones and so little effects on fatigue life.     

Delamination of Sensitized Al-Mg Alloy During Fatigue Crack Growth in Room Temperature Air

Fatigue crack growth experiments in air at 295?K (22?°C) were conducted on 6.35?mm thick plate samples of a commercial Al-Mg alloy machined from the L-T orientation. Thermal exposures for times up to 2000?hours at 343?K, 353?K, 373?K, and 448?K (70?°C, 80?°C, 100?°C, and 175?°C) produced sensitization and delamination in the S-T plane during fatigue testing, dependent on the level of thermal exposure and K _max employed. Identical tests conducted in a ??dry?? environment produced no delamination, indicating that environment may enhance the phenomena.     

A hybrid modeling approach for forecasting the volatility of S&P 500 index return

Forecasting volatility is an essential step in many financial decision makings. GARCH family of models has been extensively used in finance and economics, particularly for estimating volatility. The motivation of this study is to enhance the ability of GARCH models in forecasting the return volatility. We propose two hybrid models based on EGARCH and Artificial Neural Networks to forecast the volatility of S&P 500 index. The estimates of volatility obtained by an EGARCH model are fed forward to a Neural Network. The input to the first hybrid model is complemented by historical values of other explanatory variables. The second hybrid model takes as inputs both series of the simulated data and explanatory variables. The forecasts obtained by each of those hybrid models have been compared with those of EGARCH model in terms of closeness to the realized volatility. The computational results demonstrate that the second hybrid model provides better volatility forecasts.     

基于元建模的裂缝性气藏累计产量预测

为了预测某裂缝性气藏在18年规划周期内的累计产量,建立了3种元模型,并结合数值模拟累计产量预测结果对各模型进行了验证和对比。估算了气藏的基本参数(孔隙度、渗透率及含水饱和度)并对气藏进行了数值模拟。将气藏内某井18年内的累计产量表示为各气藏参数的函数,采用储量计算经验函数排除了研究空间内的低潜力点,并基于最大熵准则选取25个高潜力点作为设计点。建立了二次模型、乘性模型和RBF模型3种元模型,基于各模型预测了25个设计点和7个试验点处的累计产量。预测结果表明:各模型均能比较准确地预测累计产量;RBF模型的累计产量预测准确性优于另外2种模型;与数值模拟方法相比,元建模方法的计算时间大幅缩短。     

Pareto Optimal Multigene Genetic Programming for Prediction of Longitudinal Dispersion Coefficient

Water Resources Management - The longitudinal dispersion coefficient (Kx) is fundamental to modeling of pollutant and sediment transport in natural rivers, but a general expression for Kx, with...