This paper presents a new framework for object-based classification of high-resolution hyperspectral data. This multi-step framework is based on multi-resolution segmentation (MRS) and Random Forest classifier (RFC) algorithms. The first step is to determine of weights of the input features while using the object-based approach with MRS to processing such images. Given the high number of input features, an automatic method is needed for estimation of this parameter. Moreover, we used the Variable Importance (VI), one of the outputs of the RFC, to determine the importance of each image band. Then, based on this parameter and other required parameters, the image is segmented into some homogenous regions. Finally, the RFC is carried out based on the characteristics of segments for converting them into meaningful objects. The proposed method, as well as, the conventional pixel-based RFC and Support Vector Machine (SVM) method was applied to three different hyperspectral data-sets with various spectral and spatial characteristics. These data were acquired by the HyMap, the Airborne Prism Experiment (APEX), and the Compact Airborne Spectrographic Imager (CASI) hyperspectral sensors. The experimental results show that the proposed method is more consistent for land cover mapping in various areas. The overall classification accuracy (OA), obtained by the proposed method was 95.48, 86.57, and 84.29% for the HyMap, the APEX, and the CASI data-sets, respectively. Moreover, this method showed better efficiency in comparison to the spectral-based classifications because the OAs of the proposed method was 5.67 and 3.75% higher than the conventional RFC and SVM classifiers, respectively. 相似文献
This paper investigates monthly, seasonal, and annual trends in rainfall, streamflow, temperature, and humidity amounts at Urmia lake (UL) basin and analyzes the interaction between these variables and UL’s water level fluctuation during the 1971–2013 period. Two new methods including nonparametric hybrid wavelet Mann–Kendall test and ?en’s methodology have been used to determine potential trends in the variables and their dominant periods. The results showed significant decreasing trends in the water level and streamflow series, moderate decreasing trend in the rainfall and relative humidity series, and increasing trends in the observed temperature data. The 8- , 12-month, and 2-year periods were detected as the dominant periods of the variables in monthly, seasonal, and annual timescales, respectively. The results from the interaction analysis revealed that the main factor influencing the water level at UL is decreasing trend in the streamflow series. Both the monthly series of UL’s water level and the streamflow series of the stations indicated two start points of significant decreasing trend in 1973 and 1998. Furthermore, a comparative analysis among the applied methods indicated a good agreement between the results of hybrid wavelet Mann–Kendall test and ?en’s trend analyzing method. 相似文献
Maintaining the quality of surface water resources as one of the most vital water supplies has always been at the center of global concerns. A set of manifest and latent factors have yet been identified by researchers worldwide that are subject to affect the quality of surface water. Among which, the effect of land use change, due to a spatial and temporal complexity, is often not easily verifiable. The present study attempts to offer an index-based model to quantify vulnerability of surface water resources in a semi arid basin in central Iran against land use changes. For this, water quality data including Na, K, Mg, Ca, Sodium Adsorption Ratio (SAR), total anions (Sum. A), SO4, Cl, HCO3, EC, TDS, and pH were collected from hydrometric stations over a period of 26 years (1987?2013). In order to detect land use changes, the land use maps of the years 1987, 1998, 2002, 2009, and 2013 were prepared from TM satellite images using supervised classification method. At next step, changing patterns of different land uses were traced by Shanon’s Diversity Index (SHDI) as a metric of patch diversity indicating diversity and heterogeneity of a landscape over time. Relationship between the SHDI values and water quality indicators revealed the impact of land use changes on quality of surface water resources. Statistical analysis confirmed a significant relationship between nine water quality factors and water discharge in the basin during the study period. From 1998 to 2009, the greatest changes were visible in the total anions, Ca, SO4, and HCO3. These parameters along with salinity were increasing in almost all sub-basins. According to the results, conversion of poor rangelands to rain fed agriculture fields is the most apparent land use change occurred in the study area over the study period. In 1987, SHDI as an indicator of the diversity and changes in the basin, showed a significant relationship with good rangelands (R2 =–0.835). This indicates that fragmentation of the entire watershed area was initiated in 1987, which reached its peak in 2013. Generally speaking, urbanization, poor rangeland, and irrigated agriculture were recognized as three influential land uses adversely affect the water quality in the study area. 相似文献
Local and mine scale exploration models for anomaly recognition within known ore fields are discussed. Traditional geochemical exploration methods are based on multivariate statistical analysis, metallometry, vertical geochemical zonality and criteria of natural field geochemical associations, which suffer several shortcomings, including lack of a geostatistical generalised approach for separating anomalies from background. These shortcomings make the interpretation process time consuming and costly. Fuzzy set theory, fuzzy logic and neural network techniques seem very well suited for typical mining geochemistry applications. The results, obtained from applying the proposed technique to a real scenario, reveals significant improvements, comparing the results obtained from applying multivariate statistical analysis. Computationally, the introduced technique makes possible, without exploration drilling, the distinction between blind mineralisation and zone of dispersed ore mineralisation. The methodology developed in this research study has been verified by testing it on various real-world mining geochemical projects. 相似文献
Today, grouting is used as an aid in ground improvement in most civil and mining engineering projects. Groutability and grout penetration depth are among the most important issues that are considered in grouting operation. Various parameters such as in-situ stress, pore water pressure, joint geometric and geomechanical characteristics, grout properties (viscosity and yield stress) and technical factors such as grouting pressure and flow rate affect the groutability and grout penetration depth in a jointed rock mass. Knowledge of the effect of these parameters has advantages in the prediction of grouting results. Typically, cement-based grout is used in jointed rock masses. Unlike water, stable cement-based grout usually acts as a Bingham fluid. In this study the effect of important parameters on grouting process in a jointed rock mass was investigated numerically using the DEM method. In the conducted study, the problem geometry represents a horizontal section in a regularly jointed rock mass with two joint sets. The analyses results show that the grout penetration depth and intake increase as joint aperture, normal stiffness and grouting pressure increase and in-situ stress and pore water pressure decrease. Increase in joint spacing does not have any effect on the grout penetration depth but decreases the grout intake. The effect of joint orientation on grouting process is strongly dependent on in-situ stress state. On the other hand, increase in grout yield stress decreases the grout penetration depth and intake, while grout viscosity does not have any effect on maximum grout penetration depth and intake. To further investigate the above mechanisms, the grouting process conducted in Gotvand dam-Iran was simulated numerically. 相似文献
Photodegradation of PAHs in the water-soluble fraction of Kuwait crude oil in seawater was investigated under various environmental factors (temperature, light intensity, oxygen levels and presence of a sensitizer) in laboratory conditions. All factors investigated had significant effect on the degradation rates of PAHs. At 15 °C almost all PAHs optimally degraded at an oxygen level of 4 ppm. For lower molecular weight PAHs a light intensity of 500 W/m2 in the presence of the sensitizer worked well. Higher molecular weight PAHs degraded at faster rates at a light intensity 750 W/m2. At 30 °C, most of the PAHs degraded optimally at an oxygen level of 0 ppm and light intensity of 500 or 750 W/m2 in presence of the sensitizer. At 40 °C, most of PAHs degraded optimally at low oxygen concentrations (0 and 4 ppm) and a light intensity of 500 W/m2 in the presence of the sensitizer. Linear regression indicated that for most of the compounds, light intensity had the greatest effect on degradation rates. 相似文献
In nonlinear dynamic structural analysis, a suite of pulse-like ground motions is required for the performance-based design of structures near active faults. The dissimilarity in the amplitude and frequency content of the earthquake time series referred to nonstationary properties in temporal and spectral, respectively. An approach is proposed based on the nonstationary properties of the far-field records and the seismological information in an event for simulating pulse-like records. The pulse-like earthquake time history is estimated via the superposition of the residual part of the earthquake with the estimated pulse. The wavelet-based Hilbert transform is utilized to characterize the nonstationary properties, the instantaneous amplitude, and frequencies of far-field records to model residual part. The effects of near-fault and pulse are estimated based on the seismological properties of the region. The validation of the procedure is indicated by comparing simulated time-series, response spectra, and Arias intensity with recorded pulse-like records in two different earthquakes in California; the Mw 6.7 1994 Northridge and the Mw 6.5 1979 Imperial valley.
The French Transportable Laser Ranging System (FTLRS) was deployed in the calibration site of satellite radar altimeters in Corsica over the 2002 and 2005 campaigns. The paper describes the different steps of SLR data processing. The average arcs RMS obtained are about 1–2 cm for Lageos-1&;-2, Starlette and Stella satellites; it is shown that the best results of satellite orbits determination and geocentric positioning are obtained with Eigen-Grace03s gravity model. The difference of FTLRS absolute 3D positioning, between 2002 and 2005, of about 7.7 mm (i.e., 2.6 mm/yr) is less than residual errors of ITRF2005 velocities (of about 4.3 mm/yr). 相似文献