Support Vector Machine (SVM) is a popular data mining technique, and it has been widely applied in astronomical tasks, especially in stellar spectra classification. Since SVM doesn’t take the data distribution into consideration, and therefore, its classification efficiencies can’t be greatly improved. Meanwhile, SVM ignores the internal information of the training dataset, such as the within-class structure and between-class structure. In view of this, we propose a new classification algorithm-SVM based on Within-Class Scatter and Between-Class Scatter (WBS-SVM) in this paper. WBS-SVM tries to find an optimal hyperplane to separate two classes. The difference is that it incorporates minimum within-class scatter and maximum between-class scatter in Linear Discriminant Analysis (LDA) into SVM. These two scatters represent the distributions of the training dataset, and the optimization of WBS-SVM ensures the samples in the same class are as close as possible and the samples in different classes are as far as possible. Experiments on the K-, F-, G-type stellar spectra from Sloan Digital Sky Survey (SDSS), Data Release 8 show that our proposed WBS-SVM can greatly improve the classification accuracies. 相似文献
Water quality is often highly variable both in space and time, which poses challenges for modelling the more extreme concentrations. This study developed an alternative approach to predicting water quality quantiles at individual locations. We focused on river water quality data that were collected over 25 years, at 102 catchments across the State of Victoria, Australia. We analysed and modelled spatial patterns of the 10th, 25th, 50th, 75th and 90th percentiles of the concentrations of sediments, nutrients and salt, with six common constituents: total suspended solids (TSS), total phosphorus (TP), filterable reactive phosphorus (FRP), total Kjeldahl nitrogen (TKN), nitrate-nitrite (NOx), and electrical conductivity (EC). To predict the spatial variation of each quantile for each constituent, we developed statistical regression models and exhaustively searched through 50 catchment characteristics to identify the best set of predictors for that quantile. The models predict the spatial variation in individual quantiles of TSS, TKN and EC well (66%–96% spatial variation explained), while those for TP, FRP and NOx have lower performance (37%–73% spatial variation explained). The most common factors that influence the spatial variations of the different constituents and quantiles are: annual temperature, percentage of cropping land area in catchment and channel slope. The statistical models developed can be used to predict how low- and high-concentration quantiles change with landscape characteristics, and thus provide a useful tool for catchment managers to inform planning and policy making with changing climate and land use conditions. 相似文献
Remote sensing data have been widely applied to extract minerals in geologic exploration, however, in areas covered by vegetation, extracted mineral information has mostly been small targets bearing little information. In this paper, we present a new method for mineral extraction aimed at solving the difficulty of mineral identification in vegetation covered areas. The method selected six sets of spectral difference coupling between soil and plant (SVSCD). These sets have the same vegetation spectra reflectance and a maximum different reflectance of soil and mineral spectra from Hyperion image based on spectral reflectance characteristics of measured spectra. The central wavelengths of the six, selected band pairs were 2314 and 701 nm, 1699 and 721 nm, 1336 and 742 nm, 2203 and 681 nm, 2183 and 671 nm, and 2072 and 548 nm. Each data set’s reflectance was used to calculate the difference value. After band difference calculation, vegetation information was suppressed and mineral abnormal information was enhanced compared to the scatter plot of original band. Six spectral difference couplings, after vegetation inhibition, were arranged in a new data set that requires two components that have the largest eigenvalue difference from principal component analysis (PCA). The spatial geometric structure features of PC1 and PC2 was used to identify altered minerals by spectral feature fitting (SFF). The collecting rocks from the 10 points that were selected in the concentration of mineral extraction were analyzed under a high-resolution microscope to identify metal minerals and nonmetallic minerals. Results indicated that the extracted minerals were well matched with the verified samples, especially with the sample 2, 4, 5 and 8. It demonstrated that the method can effectively detect altered minerals in vegetation covered area in Hyperion image. 相似文献
In this paper, a comprehensive study on simulating the shearing behavior of frictional materials is performed. A set of two explicit equations, describing the relationship among the shear stress ratio and the distortional strain and the volumetric strain, are formulated independently. The equations contain three stress parameters and three strain parameters and another parameter representing the nonuniformity of stress and strain during softening. All the parameters have clear physical significance and can be determined experimentally. It is demonstrated that the proposed equations have the capacity of simulating the complicated shearing behavior of many types of frictional materials including geomaterials. The proposed equations are used to simulate the stress–strain behavior for 27 frictional materials with 98 tests. These materials include soft and stiff clays in both reconstituted and structured states, silicon sands and calcareous sands, silts, compacted fill materials, volcanic soils, decomposed granite soils, cemented soils (both artificially and naturally cemented), partially saturated soils, ballast, rocks, reinforced soils, tire chips, sugar, wheat, and rapeseed. It has been demonstrated that the proposed explicit constitutive equations have the capacity to capture accurately the shearing behavior of frictional materials both qualitatively and quantitatively. A study on model parameters has been performed. 相似文献
Atlantic salmon reared in recirculating aquaculture system (RAS) may lead to inappropriately high stocking density, because fish live in a limited space. Finding the suitable stocking density of Atlantic salmon reared in RAS is very important for RAS industry. In this paper, the influence of stocking density on growth and some stress related physiological factors were investigated to evaluate the effects of stocking density. The fish were reared for 220 days at five densities (A: 24 kg/m3; B: 21 kg/m3; C: 15 kg/m3; D: 9 kg/ m3 and E: 6 kg/m3 ). The results show that 30 kg/m3 might be the maximum density which RAS can afford in China. The stocking densities under 30 kg/m3 have no effect on mortality of Atlantic salmon reared in RAS. However, the specific growth rate (SGR), final weight and weight gain in the high density group were significantly lower than the lower density groups and middle density groups. Moreover, feed conversion rate (FCR) had a negative correlation with density. Plasma hormone T3 and GH showed significant decrease with the increase of the stocking density of the experiment. Furthermore, thyroid hormone (T3), GH (growth hormone) activities were decreased with stocking density increase. However, plasma cortisol, GOT (glutamic oxalacetic transaminase) and GPT (glutamic pyruvic transaminase) activities were increase with stocking density increase. And the stocking density has no effects on plasma lysozyme of Atlantic salmon reared in RAS. These investigations would also help devise efficient ways to rear adult Atlantic salmon in China and may, in a way, help spread salmon mariculture in China.