A Multiple SVM System for Classification of Hyperspectral Remote Sensing Data

With recent technological advances in remote sensing sensors and systems, very high-dimensional hyperspectral data are available for a better discrimination among different complex land-cover classes. However, the large number of spectral bands, but limited availability of training samples creates the problem of Hughes phenomenon or ‘curse of dimensionality’ in hyperspectral data sets. Moreover, these high numbers of bands are usually highly correlated. Because of these complexities of hyperspectral data, traditional classification strategies have often limited performance in classification of hyperspectral imagery. Referring to the limitation of single classifier in these situations, Multiple Classifier Systems (MCS) may have better performance than single classifier. This paper presents a new method for classification of hyperspectral data based on a band clustering strategy through a multiple Support Vector Machine system. The proposed method uses the band grouping process based on a modified mutual information strategy to split data into few band groups. After the band grouping step, the proposed algorithm aims at benefiting from the capabilities of SVM as classification method. So, the proposed approach applies SVM on each band group that is produced in a previous step. Finally, Naive Bayes (NB) as a classifier fusion method combines decisions of SVM classifiers. Experimental results on two common hyperspectral data sets show that the proposed method improves the classification accuracy in comparison with the standard SVM on entire bands of data and feature selection methods.     

Palynomorph stratigraphy,palynofacies and organic geochemistry assessments for hydrocarbon generation of Ratawi Formation,Iraq

Seventy-two core and cutting samples of the Ratawi Formation from selected wells of central and southern Iraq in Mesopotamian Foredeep Basin are analysed for their sedimentary organic matters. Dinoflagellates, spores and pollen are extracted by palynological techniques from these rocks. Accordingly, Hauterivian and late Valanginian ages are suggested for their span of depositional time. These palynomorphs with other organic matter constituents, such as foraminifer’s linings, bacteria and fungi, are used to delineate three palynofacies types that explain organic matter accumulation sites and their ability to generate hydrocarbons. Palaeoenvironments of these sites were mainly suboxic to anoxic with deposition of inshore and neritic marine environments especially for palynofacies type 2. Total organic matters of up to 1.75 total organic carbon (TOC) wt.% and early mature stage of up to 3.7 TAI based on the brown colour of the spore species Cyathidites australis and Gleichenidites senonicus with mottled interconnected amorphous organic matter are used for hydrocarbon generation assessment from this formation. On the other hand, these rock samples are processed with Rock-Eval pyrolysis. Outcomes and data calculations of these analyses are plotted on diagrams of kerogen types and hydrocarbon potential. Theses organic matter have reached the mature stage of up to T _max?=?438 °C, hydrogen index of up to 600 mg hydrocarbons for each gram of TOC wt.% and mainly low TOC (0.50–1.55). Accordingly, this formation could generate fair quantities of hydrocarbons in Baghdad oil field and Basrah oil fields. Organic matters of this formation in the fields of Euphrates subzone extends from Hilla to Nasiriyah cities have not reached mature stage and hence not generated hydrocarbons from the Ratawi Formation. Software 1D PetroMod basin modelling of the Ratawi Formation has confirmed this approach of hydrocarbon generation with 100 % transformations of the intended organic matters to generate hydrocarbons to oil are performed in especially oil fields of East Baghdad, West Qurna and Majnoon while oil fields Ratawi and Subba had performed 80–95 % transformation to oil and hence end oil generation had charged partly the Tertiary traps that formed during the Alpine Orogeny. Oil fields of Nasiriyah and Kifle had performed least transformation ratio of about 10–20 % transformation to oil, and hence, most of the present oil in this field is migrated from eastern side of the Mesopotamian Foredeep Basin that hold higher maturation level.     

Effective prediction of scour downstream of ski-jump buckets using artificial neural networks

The main objective of this research was to analyze and quantify the uncertainty of artificial neural network in prediction of scour downstream ski-jump buckets. Hence, at first, three artificial neural network models were developed to predict depth, length, and width of scour hole. Then, Monte-Carlo simulation was applied in the estimates of artificial neural network modeling procedure. The uncertainties were quantified by means of two criteria: 95 percent prediction uncertainty and d-factor. The results of the artificial neural network models showed superior performance of it in comparison with some empirical formulas because of higher correlation coefficient (R ² > 0.95) and lower error (RMSE < 1.63). The obtained result from uncertainty analysis of the models revealed the satisfactory performance of them. In this procedure it was clarified that the artificial neural network model for length prediction was more reliable than the others with d-factor and 95 percent prediction uncertainty equal to 2.53 and 92, respectively.     

The effects of agricultural practice and land-use on the distribution and origin of some potentially toxic metals in the soils of Golestan province, Iran

Soil samples were collected from the agricultural lands of Golestan province, north of Iran and analyzed for 24 elements including eight toxic metals of As, Cd, Co, Cr, Cu, Pb, Se and Zn. Electrical conductivity, pH, organic matter, soil texture, calcium carbonate content as well as soil cation exchange capacity were also determined. The possible sources of metals are identified with multivariate analysis such as correlation analysis, principal component analysis (PCA), and cluster analysis. In addition, enrichment factors were used to quantitatively evaluate the influences of agricultural practice on metal loads to the surface soils. The PCA and cluster analysis studies revealed that natural geochemical background are the main source of most elements including Al, Co, Cr, Cs, Cu, Fe, K, Li, Ni, Pb, V and Zn in the arable soils of the province (more than 90 %), however, those soils which have been developed on the mafic and metamorphic rocks were considerably contributed on metal concentration (43 %). Calcium and Sr were constituents of calcareous rocks and Na and S were mainly controlled by saline soils in the north of the province. Loess deposits was also accounting for high levels of selenium concentration. Phosphorous was mostly related to application of P-fertilizers and organophosphate pesticides. The comparison of metal load and enrichment factor for dry and irrigated farmlands showed that Cd, Co, Pb, Se and Zn had higher concentrations in the irrigated lands where considerable amounts of agrochemicals had been applied. However, it also found that proximity of arable lands to urban and industrial areas resulted in higher Pb and Cd values in the irrigated agricultural sources relative to dry ones.     

Estimation of uniaxial compressive strength based on P-wave and Schmidt hammer rebound using statistical method

Preparing high-quality samples, which can fulfill testing standards, from weak and block-in-matrix conglomerate for laboratory tests, is a big challenge in engineering projects. Hence, using indirect methods seems to be indispensable for determination uniaxial compressive strength (UCS). The main objective of this study is to estimate the relation between sonic velocity (Vp), Schmidt hammer rebound number (SCH) and UCS. For this reason, some samples of weak conglomeratic rock were collected from two different sites of dam in Iran (Bakhtiari and Hezardareh Formations). In order to evaluate the correlation, the measured and predicted values utilizing simple and multivariate regression techniques were examined. To control the performance of the proposed equation, root mean square error (RMSE) and value accounts for (VAF%) were determined. The VAF% and RMSE indices were computed as 94.34 and 1.56 for the relation between Vp and UCS from simple regression model. These were 94.39 and 1.6 between SCH and UCS, while these were 97.24 and 1.34 for uniaxial compressive strengths obtained from multivariate regression model.     

Effect of Pile Driving on Ground Vibration in Clay Soil: Numerical and Experimental Study

Geotechnical and Geological Engineering - In this study, peak particle velocity (PPV) values for driving three piles with diameters of 40&nbsp;cm, 50&nbsp;cm, and 70&nbsp;cm in a clayey...