Class imbalance in unsupervised change detection – A diagnostic analysis from urban remote sensing

Automatic monitoring of changes on the Earth’s surface is an intrinsic capability and simultaneously a persistent methodological challenge in remote sensing, especially regarding imagery with very-high spatial resolution (VHR) and complex urban environments. In order to enable a high level of automatization, the change detection problem is solved in an unsupervised way to alleviate efforts associated with collection of properly encoded prior knowledge. In this context, this paper systematically investigates the nature and effects of class distribution and class imbalance in an unsupervised binary change detection application based on VHR imagery over urban areas. For this purpose, a diagnostic framework for sensitivity analysis of a large range of possible degrees of class imbalance is presented, which is of particular importance with respect to unsupervised approaches where the content of images and thus the occurrence and the distribution of classes are generally unknown a priori. Furthermore, this framework can serve as a general technique to evaluate model transferability in any two-class classification problem. The applied change detection approach is based on object-based difference features calculated from VHR imagery and subsequent unsupervised two-class clustering using k‐means, genetic k-means and self-organizing map (SOM) clustering. The results from two test sites with different structural characteristics of the built environment demonstrated that classification performance is generally worse in imbalanced class distribution settings while best results were reached in balanced or close to balanced situations. Regarding suitable accuracy measures for evaluating model performance in imbalanced settings, this study revealed that the Kappa statistics show significant response to class distribution while the true skill statistic was widely insensitive to imbalanced classes. In general, the genetic k-means clustering algorithm achieved the most robust results with respect to class imbalance while the SOM clustering exhibited a distinct optimization towards a balanced distribution of classes.     

支持向量机组合方法在砂泥岩储层岩性识别中的应用

岩性是表征储油物性、建立各类地质模型的重要参数,其与测井参数的函数关系很复杂。基于支持向量机一类对一类法对岩性识别分类精度不高的现状,结合二叉树法初始分类精度较高、分类速度快等优点,提出了一种新的组合方法——一类对一类法与二叉树法的结合应用。该方法对样本数据较少的类别设置权重系数,减少样本不平衡的影响,并利用几何平均准确率作为评价岩心识别效果的指标,其对岩性的分类效果远优于单一方法。具体步骤为:首先对不均衡的样本设置相应的权重系数,然后利用二叉树法将易于与砂泥岩区分的灰岩区分开来,再利用一对一分类法将剩下的砂泥岩样本进行分类。运用此方法对某油田测井数据进行岩性分类,分类的整体准确率以及几何平均准确率均有很大的提高。     

SURFACE IMBALANCE ENERGY CALCULATED AND ANALYZED WITH THE DATA OF EBEX-2000*

Mainly.three methods have been developed to calculate turbulence heat flux.They are eddy covariance method,Bowen ratio/energy balance method and aerodynamic method.In this paper, all the three methods have been used to calculate sensible heat flux,latent heat flux and imbalance energy near the surface with the experiment data of EBEX-2000.Then comparisons of the three methods and some possible explanations of the surface imbalance energy are given.     

Evolution of the magnetic field imbalance on the Sun

The time and latitude change of the flux and rotation of magnetic-field imbalance structures with various strengths has been determined from observations at the Kitt-Peak observatory for 26 years. The regularities revealed during the work allow this change to be explained as follows. The structure of the imbalance of the magnetic field of a particular strength emerges at the photosphere surface while possessing a rotation typical for the area of this structure formation. After this, the structure begins to drift along the meridian (toward the pole or toward the equator) while rotating at the same velocity and occupying several interval of latitudes. Having displaced to the poles from the emerging latitude by about 20° (or more, depending on the rotation period), structures that have a certain significant period cease to exist as a whole, giving rise to other structures with other significant rotation periods. From here it follows that the differential rotation of the layers responsible for forming the imbalance structures of fields with various strengths can be determined from the dependence of the rotation period on the latitude of the emergence of the imbalance structure.     

Temporal and spatial differences and imbalance of China’s urbanization development during 1950–2006

A better understanding of the regional disparity and imbalance characteristics of China’s urbanization development is the important premise for constituting correct policy and strategy and promoting the healthy and sustainable development of urbanization in the 21st century. The regional differences of China’s urbanization level have close relations with natural conditions of landform and climate etc., the urbanization level reduces with the elevation of topography and decrease of precipitation. According to the statistical data set of urbanization in 1950–2006, the temporal change course of inter-provincial disparity of Chinese urbanization level since the founding of New China in 1949 was studied, and then the inter- regional and intra-regional disparities of urbanization development were analyzed by the Theil index and its nested decomposition method, to grasp the dynamic change of spatial disparities of China’s urbanization level on the whole. Using the imbalance index model, the imbalance status of urban population distribution relative to total population, grain output, total agricultural output value, gross output value of industry, tertiary industrial output value as well as gross regional product was discussed, to hold the balance characteristics of urbanization development relative to the regional development conditions from the macroscopic scales.     

Temporal and spatial differences and imbalance of China＇s urbanization development during 1950-2006

A better understanding of the regional disparity and imbalance characteristics of China’s urbanization development is the important premise for constituting correct policy and strategy and promoting the healthy and sustainable development of urbanization in the 21st century. The regional differences of China’s urbanization level have close relations with natural conditions of landform and climate etc., the urbanization level reduces with the elevation of topography and decrease of precipitation. According to the statistical data set of urbanization in 1950–2006, the temporal change course of inter-provincial disparity of Chinese urbanization level since the founding of New China in 1949 was studied, and then the inter-regional and intra-regional disparities of urbanization development were analyzed by the Theil index and its nested decomposition method, to grasp the dynamic change of spatial disparities of China’s urbanization level on the whole. Using the imbalance index model, the imbalance status of urban population distribution relative to total population, grain output, total agricultural output value, gross output value of industry, tertiary industrial output value as well as gross regional product was discussed, to hold the balance characteristics of urbanization development relative to the regional development conditions from the macroscopic scales.     

Studies on the transfer of mineral nutrients from feed, water, soil and plants to buffaloes under arid environments

Studies were conducted at a Livestock Farm in an arid region of Pakistan, to determine the uptake of mineral nutrients from the food of buffaloes. Samples of feed, water, forage, soil as well as blood, milk, urine, and faeces of the animals were collected and analysed for K⁺, Ca²⁺, and Fe²⁺ during summer and winter. The results showed that the concentrations of K⁺ and Ca²⁺ in plants were low, whereas that of Fe²⁺ was high. In animals, K⁺ was not found in the milk in significant amounts, rather it was excreted in the urine. The calcium taken in by the animals was used for the synthesis of milk, as a result of which low Ca²⁺ concentrations were detected in urine. Mineral nutrient concentrations (K⁺ and Ca²⁺) in forage plants and water were lower than the optimum requirement of the animals, as a result of which blood plasma contained lower amounts of minerals (K⁺ and Fe²⁺) than that of the normal animals or critical limits. From the findings of the present study, it can be concluded that the buffaloes raised at the livestock farm had a severe deficiency of mineral nutrients, particularly K⁺ in lactating and Fe²⁺ in both lactating and non-lactating animals. Feed supplements, that can overcome these deficiencies, should be used for the optimum growth, milk production and resistance to diseases.     

中国区域经济时空动态不平衡发展分析

区域经济发展是受区域内外的多重因素影响的 ,对区域经济发展的空间不平衡动态规律的研究是国内外学者研究的重要课题。本文应用区域经济空间动态发展模型对中国 1978～ 1998年的区域经济生产率增长率进行分析 ,寻找影响其增长的区域内外因素及变动规律。考虑到中国的本土化特色 ,在模型中引入开放优惠政策和区位优势变量 ,回归结果得到显著提高。研究表明 ,在过去的 2 0年间中国区域经济发展与国外 (如欧盟 )的发展规律不完全相同 ,具有中国特色。尽管各地区间的经济发展存在差异 ,但总体上中国经济发展满足一定的模式。     

区域发展的空间失衡模式与状态评估——以江苏省为例

从收入增长和生态改善的空间均衡角度来看,区域发展空间失衡是空间开发状态与其区位供给能力不相匹配,空间的经济社会活动没有按照开发供给的比较优势进行地区分工配置。当区域开发规模及程度超过其可能的供给容量视为开发过度;若没有充分利用供给容量,则为开发不足。以江苏省县(县级市) 单元为例,采用开发强度与反映空间供给能力的生态重要性指数、环境敏感性指数、规模集聚指数、资源保障指数、运输经济指数的相关关系以及协调度,可以判断和评价空间失衡的状态。同时,资源环境价格市场不完善,政府调控要素导向不合理以及开发导向的财税和政绩考核体系等安排,都是导致空间失衡的制度原因。     

材质和倾角属性筛选的光伏屋顶提取

针对城市光伏屋顶类型多样致使样本不平衡,高空间分辨率卫星影像光伏屋顶提取问题,提出了一种材质和倾角属性筛选的方法。通过选取天津市南开区、红桥区、和平区、河东区卫星影像,利用标注于光伏屋顶的材质和倾角两种属性筛选出样本均衡的数据集,借助TensorFlow框架下集成的U-Net算法并合理设置参数,得到了光伏屋顶提取结果,并与未利用材质和倾角两种属性筛选情况下的光伏屋顶提取结果进行对比。试验结果表明：本文方法提取精度较高,尤其是对于研究区域不常见的光伏屋顶,能够提取出更为完整准确且边界清晰的结果。