An evaluation of Guided Regularized Random Forest for classification and regression tasks in remote sensing

New Earth observation missions and technologies are delivering large amounts of data. Processing this data requires developing and evaluating novel dimensionality reduction approaches to identify the most informative features for classification and regression tasks. Here we present an exhaustive evaluation of Guided Regularized Random Forest (GRRF), a feature selection method based on Random Forest. GRRF does not require fixing a priori the number of features to be selected or setting a threshold of the feature importance. Moreover, the use of regularization ensures that features selected by GRRF are non-redundant and representative. Our experiments based on various kinds of remote sensing images, show that GRRF selected features provides similar results to those obtained when using all the available features. However, the comparison between GRRF and standard random forest features shows substantial differences: in classification, the mean overall accuracy increases by almost 6% and, in regression, the decrease in RMSE almost reaches 2%. These results demonstrate the potential of GRRF for remote sensing image classification and regression. Especially in the context of increasingly large geodatabases that challenge the application of traditional methods.     

GDJ人卫目视观测数据采集和处理系统的研制

本文介绍了ＧＤＪ跟踪打印经纬仪在人造卫星的观测中如何应用计算机对多圈卫星的目视观测资料进行实时采集、处理等工作过程的原理和方法。     

A critical state model for sands dependent on stress and density

An elastoplastic model for sands is presented in this paper, which can describe stress–strain behaviour dependent on mean effective stress level and void ratio. The main features of the proposed model are: (a) a new state parameter, which is dependent on the initial void ratio and initial mean stress, is proposed and applied to the yield function in order to predict the plastic deformation for very loose sands; and (b) another new state parameter, which is used to determine the peak strength and describe the critical state behaviour of sands during shearing, is proposed in order to predict simply negative/positive dilatancy and the hardening/softening behaviour of medium or dense sands. In addition, the proposed model can also predict the stress–strain behaviour of sands under three-dimensional stress conditions by using a transformed stress tensor instead of ordinary stress tensor. Copyright © 2004 John Wiley & Sons, Ltd.     

On the practice of estimating fractal dimension

Coastlines epitomize deterministic fractals and fractal (Hausdorff-Besicovitch) dimensions; a divider [compass] method can be used to calculate fractal dimensions for these features. Noise models are used to develop another notion of fractals, a stochastic one. Spectral and variogram methods are used to estimate fractal dimensions for stochastic fractals. When estimating fractal dimension, the objective of the analysis must be consistent with the method chosen for fractal dimension calculation. Spectal and variogram methods yield fractal dimensions which indicate the similarity of the feature under study to noise (e.g., Brownian noise). A divider measurement method yields a fractal dimension which is a measure of complexity of shape.     

Anhydrite pseudomorphs and the origin of stratiform Cu–Co ores in the Katangan Copperbelt (Democratic Republic of Congo)

The stratiform Cu–Co ore mineralisation in the Katangan Copperbelt consists of dispersed sulphides and sulphides in nodules and lenses, which are often pseudomorphs after evaporites. Two types of pseudomorphs can be distinguished in the nodules and lenses. In type 1 examples, dolomite precipitated first and was subsequently replaced by Cu–Co sulphides and authigenic quartz, whereas in type 2 examples, authigenic quartz and Cu–Co sulphides precipitated prior to dolomite and are coarse-grained. The sulphur isotopic composition of the copper–cobalt sulphides in the type 1 pseudomorphs is between −10.3 and 3.1‰ relative to the Vienna Canyon Diablo Troilite, indicating that the sulphide component was derived from bacterial sulphate reduction (BSR). The generation of during this process caused the precipitation and replacement of anhydrite by dolomite. A second product of BSR is the generation of H₂S, resulting in the precipitation of Cu–Co sulphides from the mineralising fluids. Initial sulphide precipitation occurred along the rim of the pseudomorphs and continued towards the core. Precipitation of authigenic quartz was most likely induced by a pH decrease during sulphide precipitation. Fluid inclusion data from quartz indicate the presence of a high-salinity (8–18 eq. wt.% NaCl) fluid, possibly derived from evaporated seawater which migrated through the deep subsurface. ⁸⁷Sr/⁸⁶Sr ratios of dolomite in type 1 nodules range between 0.71012 and 0.73576, significantly more radiogenic than the strontium isotopic composition of Neoproterozoic marine carbonates (⁸⁷Sr/⁸⁶Sr = 0.7056–0.7087). This suggests intense interaction with siliciclastic sedimentary rocks and/or the granitic basement. The low carbon isotopic composition of the dolomite in the pseudomorphs (−7.02 and −9.93‰ relative to the Vienna Pee Dee Belemnite, V-PDB) compared to the host rock dolomite (−4.90 and +1.31‰ V-PDB) resulted from the oxidation of organic matter during BSR.     

Influence of Structural Non-Stationarity of Surface Roughness on Morphological Characterization and Mechanical Deformation of Rock Joints

Summary Structural non-stationarity of surface roughness affects accurate morphological characterization as well as mechanical behaviour of rock joints at the laboratory scale using samples with a size below the stationarity threshold. In this paper, the effect of structural non-stationarity of surface roughness is investigated by studying the scale dependence of surface roughness and mechanical behaviour of rock joints. The results show that the structural non-stationarity mainly affects the accurate characterization of the surface roughness of the fracture samples. It also controls the amount and location of the contact areas during shear tests, which in turn affects the mechanical properties and asperity degradation of the samples. It is concluded that for accurate determination of the morphological and mechanical properties of rock joints at laboratory and field scales, samples with size equal to or larger than the stationarity threshold are required. Author’s address: Nader Fardin, Rock Mechanics Group, Department of Mining Engineering, Faculty of Engineering, University of Tehran, P.O. Box: 11365/4563, Tehran, Iran     

Implementing low-cost landslide risk reduction: a pilot study in unplanned housing areas of the Caribbean

Landslides pose a serious physical and environmental threat to vulnerable communities living in areas of unplanned housing on steep slopes in the Caribbean. Some of these communities have, in the past, had to be relocated, at costs of millions of dollars, because of major slides triggered by tropical storm rainfall. Even so, evidence shows that: (1) risk reduction is a marginal activity; (2) there has been minimal uptake of hazard maps and vulnerability assessments and (3) there is little on-the-ground delivery of construction for risk reduction. This article directly addresses these issues by developing a low-cost approach to the identification of the potential pore pressure changes that trigger such slides we seek to address these three commentaries directly. A complex 45–60° slope site in St Lucia, West Indies was selected as a pilot for a modelling approach that uses numerical models (FLAC and CHASM) to verify the need for surface water management to effectively reduce landslide risk. Following the model confirmation, a series of drains were designed and constructed at the site. Post-construction evidence indicates the methodology to be sound, in that the site was stable in subsequent 1-in-1 to 1-in-4 year rainfall events. A critical feature of the approach is that it is community-based from data acquisition through to community members participating in construction.     

固定式海洋平台利用TLD的减震研究

本文研究固定式海洋平台利用调谐液体阻尼器（ＴｕｎｅｄＬｉｑｕｉｄＤａｍｐｅｒ，简称ＴＬＤ）减小地震反应的方法。首先探讨了调谐频率比、激磁频率比对减震效果的影响，在此基础上建立了ＴＬＤ——平台系统的力学模型和运动方程，通过数值计算验证了该方法的有效性     

Benthic mineralization rates at two locations in the southern North Sea

Benthic oxygen uptake, sulphate reduction and benthic bacterial production were measured at two contrasting locations in the southern North Sea: the shallow and turbulent Broad Fourteens area in the Southern Bight, and the deeper Oyster Grounds, a deposition area, where thermohaline stratification occurs during summer. Oxygen uptake and sulphate reduction showed a clear seasonal pattern in the Broad Fourteens area, indicating a supply of carbon to the benthic system that is closely related to the standing stock of carbon in the water column. This close benthic-pelagic coupling is probably due to the influence of the tide in this part of the North Sea, which keeps the water column permanently mixed. At the Oyster Grounds, no seasonal pattern was observed. Peaks in oxygen uptake and sulphate reduction were found in winter. Irregularly occurring events, such as storms and fishery-related activities, are likely to affect the benthic mineralization patterns in this area. Annual benthic carbon mineralization rates estimated from oxygen uptake rates were 44 gC·m⁻² at the Broad Fourteens, and 131 gC·m⁻² at the Oyster Grounds, of which 26 and 28%, respectively, could be attributed to sulphate reduction (assuming an annual sulphide reoxidation rate of 100%). Although sulphate reduction rates in the southern North Sea are higher than previously suggested, aerobic respiration is the most important pathway for benthic carbon mineralization at the stations visited. Production rates of benthic bacterial carbon measured with labelled leucine were much higher than carbon mineralization rates based on oxygen uptake or sulphate reduction. This may either imply a very high bacterial carbon conversion efficiency, or point to shortcomings in the accuracy of the techniques. A critical evaluation of the techniques is recommended.     

Fractal Determination of Coastal Types Based on GIS--A Case Study of China

The probability of fractal determination of coastal types based on GIS is preliminarily discussed with China as an example. Finally, some significant conclusions are drawn: (1) The fractal dimension of coastline of the bedrock coast is larger than that of the plain coast on the same scale map; (2) As far as the bedrock coast is concerned, the larger fractal dimension of coastline of the bedrock coast on the same scale map indicates that the bedrock coast is probably not typical; (3) As far as the plain coast is concerned, the smaller fractal dimension of coastline of the plain coast on the same scale map indicates that it is probably the silt plain coast; (4) The different substantial compositions affect the fractal dimensions of coastlines of different coastal types. In general, the coast which lies in the north of the Hangzhou Bay consists of sand mainly, its surface is flat, and it is connected with the coastal plain, its landform is broad shoal, its total change is comparatively homogenous in the tidal dynamic process, and thus, the relatively smaller fractal dimension of coastline results from this. For the bedrock coast, there is more bedrock, the coastline is comparatively smooth and straight, being affected by the faults and ocean dynamic process, which result in the larger fractal dimension.