Automated reconstruction of building objects from aerial images is a complex problem due to the diversity of buildings as well as noise and low contrast of images, which are the results of distant photography, atmospheric effects and poor illumination. In this paper, a semi-automated approach to the reconstruction of parametric building models from aerial images is presented, which works with line segments extracted from the image. The model is selected interactively from a library of parametric models. A perceptual grouping technique is used to select the most significant image lines in terms of relations such as proximity and parallelism. Model lines are searched for the same relations as in the grouped image lines, and the corresponding lines undergo a matching procedure, which determines whether or not a match can be found between the given model and image lines. An experiment with aerial images of flat-roof and gable-roof buildings is shown and its results indicate the robustness and efficiency of the proposed approach. 相似文献
Geostatistical simulation aims at reproducing the variability of the real underlying phenomena. When nonlinear features or large-range connectivity is present, the traditional variogram-based simulation approaches do not provide good reproduction of those features. Connectivity of high and low values is often critical for grades in a mineral deposit. Multiple-point statistics can help to characterize these features. The use of multiple-point statistics in geostatistical simulation was proposed more than 10 years ago, on the basis of the use of training images to extract the statistics. This paper proposes the use of multiple-point statistics extracted from actual data. A method is developed to simulate continuous variables. The indicator kriging probabilities used in sequential indicator simulation are modified by probabilities extracted from multiple-point configurations. The correction is done under the assumption of conditional independence. The practical implementation of the method is illustrated with data from a porphyry copper mine. 相似文献
Trend modelling is an important part of natural resource characterization. A common approach to account for a variable with a trend is to decompose it into a relatively smoothly varying trend and a more variable residual component. Then, the residuals are stochastically modelled independent of the trend. This decomposition can result in values outside the plausible range of variability, such as grades below zero or ratios that exceed 1.0. We transform the residuals conditional to the trend component to explicitly remove these complex features prior to geostatistical modelling. Back transformation of the modelled residual values allows the complex relations to be reproduced. A petroleum-related application shows the robustness of the proposed transformation. Furthermore, a mining application shows that when this conditional transformation is applied to the original variable, instead of the residual, simulated values are assured to be nonnegative. 相似文献
Rock mass classification is analogous to multi-feature pattern recognition problem. The objective is to assign a rock mass to one of the pre-defined classes using a given set of criteria. This process involves a number of subjective uncertainties stemming from: (a) qualitative (linguistic) criteria; (b) sharp class boundaries; (c) fixed rating (or weight) scales; and (d) variable input reliability. Fuzzy set theory enables a soft approach to account for these uncertainties by allowing the expert to participate in this process in several ways. Hence, this study was designed to investigate the earlier fuzzy rock mass classification attempts and to devise improved methodologies to utilize the theory more accurately and efficiently. As in the earlier studies, the Rock Mass Rating (RMR) system was adopted as a reference conventional classification system because of its simple linear aggregation.
The proposed classification approach is based on the concept of partial fuzzy sets representing the variable importance or recognition power of each criterion in the universal domain of rock mass quality. The method enables one to evaluate rock mass quality using any set of criteria, and it is easy to implement. To reduce uncertainties due to project- and lithology-dependent variations, partial membership functions were formulated considering shallow (<200 m) tunneling in granitic rock masses. This facilitated a detailed expression of the variations in the classification power of each criterion along the corresponding universal domains. The binary relationship tables generated using these functions were processed not to derive a single class but rather to plot criterion contribution trends (stacked area graphs) and belief surface contours, which proved to be very satisfactory in difficult decision situations. Four input scenarios were selected to demonstrate the efficiency of the proposed approach in different situations and with reference to the earlier approaches. 相似文献
Current land-use classifications used to assess urbanization effects on stream water quality date back to the 1980s when limited information was available to characterize watershed attributes that mediate non-point source pollution. With high resolution remote sensing and widely used GIS tools, there has been a vast increase in the availability and precision of geospatial data of built environments. In this study, we leverage geospatial data to expand the characterization of developed landscapes and create a typology that allows us to better understand the impact of complex developed landscapes across the rural to urban gradient. We assess the ability of the developed landscape typology to reveal patterns in stream water chemistry previously undetected by traditional land-cover based classification. We examine the distribution of land-cover, infrastructure, topography and geology across 3876 National Hydrography Dataset Plus catchments in the Piedmont region of North Carolina, USA. From this dataset, we generate metrics to evaluate the abundance, density and position of landscape features relative to streams, catchment outlets and topographic wetness metrics. While impervious surfaces are a key distinguishing feature of the urban landscape, sanitary infrastructure, population density and geology are better predictors of baseflow stream water chemistry. Unsupervised clustering was used to generate a distinct developed landscape typology based on the expanded, high-resolution landscape feature information. Using stream chemistry data from 37 developed headwater catchments, we compared the baseflow water chemistry grouped by traditional land-cover based classes of urbanization (rural, low, medium and high density) to our composition and structure-based classification (a nine-class typology). The typology based on 22 metrics of developed landscape composition and structure explained over 50% of the variation in NO3−-N, TDN, DOC, Cl−, and Br− concentration, while the ISC-based classification only significantly explained 23% of the variation in TDN. These results demonstrate the importance of infrastructure, population and geology in defining developed landscapes and improving discrete classes for water management. 相似文献