点状地名信息的加权泰森多边形检索法

鉴于地名库中多数地名仅以中心点坐标形式记录其空间位置信息,缺乏其空间范围的具体描述,造成地名检索中的片面性和局限性,本文在深入分析地名及其空间属性、空间关系的基础上,考虑同类型地名的不同性质在检索中的重要作用,利用泰森多边形在地名边界近似中的优势,提出了针对点状地名信息的加权泰森多边形检索法,依据同类型地名的面积属性作为权重指标构建泰森多边形,以近似表达地名的空间范围,进而描述地名间的各类空间关系;并从拓扑关系、方位关系与距离关系3个方面给出了具体公式以计算地名间的空间相似性;最后以行政区划边界近似与检索为例验证该方法。试验结果表明,该方法可较好地近似表达地名空间范围及空间关系,增强了检索词与地理信息资源在空间位置上的相似性度量,检索结果优于传统方法。     

基于条件随机场的中文地名识别方法

在互联网迅速发展的现代化信息社会,大量地理信息都以非结构化的文本形式存在,而地名识别是挖掘这些地理信息的重要基础。目前已有的地名识别方法主要是从自然语言处理的角度来实现,并没有充分考虑到地名的构成和使用习惯等特征,造成识别率偏低或过拟合等问题。本文引入语言学相关知识,分析中文地名用字特征,在传统的地名专名+通名的结构上,更细致地划分地名的词素类型,总结归纳各词素类型的特征,将这些特征融入条件随机场的方法中,使地名识别问题转化为序列标注问题。并根据中文地名的特征,制定形式化规则,设计基于字的标注规范。在此基础上,设计中文地名特征模板,通过条件随机场模型训练和预测,识别自然语言文本中的中文地名。采用170万字的人民日报标注语料进行实验验证,结果表明本文方法对中文地名识别的召回率、准确率和F值分别达到92.69%、96.73%和94.67%,优于已有研究成果,能为地理信息科学领域的研究和应用提供更有效的地名服务。     

A points of interest matching method using a multivariate weighting function with gradient descent optimization

Volunteered geographic information contains abundant valuable data, which can be applied to various spatiotemporal geographical analyses. While the useful information may be distributed in different, low‐quality data sources, this issue can be solved by data integration. Generally, the primary task of integration is data matching. Unfortunately, due to the complexity and irregularities of multi‐source data, existing studies have found it difficult to efficiently establish the correspondence between different sources. Therefore, we present a multi‐stage method to match multi‐source data using points of interest. A spatial filter is constructed to obtain candidate sets for geographical entities. The weights of non‐spatial characteristics are examined by a machine learning‐related algorithm with artificially labeled random samples. A case study on Fuzhou reveals that an average of 95% of instances are accurately matched. Thus, our study provides a novel solution for researchers who are engaged in data mining and related work to accurately match multi‐source data via knowledge obtained by the idea and methods of machine learning.     

基于伪语义相似度模型的中文地址匹配方法

地址匹配中,由于传统相似度模型受字符重叠数影响大,在处理简写、缩写地址要素单元时,错误匹配问题突出;深度学习方法需要大量样本支撑,但庞大的数据量和多样的形式,导致生成样本的成本过高。为解决上述问题,本文首先应用基于条件随机场和双向长短时记忆神经网络的模型,对地址进行分词;然后通过建立一种伪语义相似度,对地址要素进行分级匹配。通过对公安业务中地址数据进行测试,在对缩写、简写等不规范地址描述方面,本文模型能较理想地完成任务,各参考指标均高于0.9。     

高分辨率遥感影像解译中的机器学习范式

高分辨率遥感影像解译是遥感信息处理领域的研究热点之一,在遥感大数据知识挖掘与智能化分析中起着至关重要的作用,具有重要的民用和军事应用价值.传统的高分辨率遥感影像解译通常采用人工目视解译方式,费时费力且精度低.所以,如何自动、高效地实现高分辨率遥感影像解译是亟待解决的问题.近年来,随着人工智能技术的飞速发展,采用机器学习...     

基于GIS的榆林市乡镇地名分布特征研究

利用不同的网络公开地名数据源进行数据融合,获得榆林市乡镇级居民地地名数据,在此基础上,对地名数据进行统计分类,引入地名密度和地名点距离的指标,运用GIS的分析方法,对榆林市乡镇级地名数据进行景观分析。研究得出:榆林乡镇地名从东南向西北部递减,并且呈现出聚集的分布特征;榆林市的自然人文特征直接体现在其主要地名分类的分布上,山水类地名的分布反映了榆林西北部沙漠干旱,东南部黄土丘陵、河流较密的自然特点,姓氏、军事类地名的分布则反映了榆林市的移民文化、多民族混居的民族特色,以及古代军事价值高的特点。     

软件模块故障倾向预测方法研究

研究了在区分故障严重程度下的软件模块故障倾向预测方法,将故障分为高严重程度和低严重程度两种类型,用统计分析和机器学习方法分析静态代码度量与故障倾向之间的关系。以公开和私有两种类型的失效数据集作为实验数据,分析发现,故障的严重程度影响预测性能,预测不同严重程度的故障需要选择不同的度量和分类模型,预测低严重程度故障的性能好于预测高严重程度故障的性能。     

采用主成分分析的面状居民地匹配指标简化方法

现有多源居民地匹配中存在众多的面要素度量指标,若全部进行考虑,则增加了匹配的复杂性;若只考虑部分指标,则可能造成匹配信息的缺失,影响匹配结果。针对这一问题,本文提出一种采用主成分分析方法的面状居民地匹配方法。借鉴主成分分析法中降维的思想,对居民地各项度量指标进行定性定量分析,通过科学计算确定面要素匹配综合指标,用较少的新指标代替原来较多的相似性指标,进而根据获得的整体相似性评价指标进行居民地匹配。实验分析表明,本文方法简化了匹配过程中众多的相似性指标,降低了匹配复杂性和不确定性,避免了各相似权值确定较为随意的问题,有效提高了匹配效率和正确率。     

无监督密集匹配特征提取网络性能分析

随着人工智能的发展,基于深度学习的有监督密集匹配方法在虚拟、室内及驾驶等近景数据集上取得了不错的表现。针对航空影像密集匹配标签数据获取困难的问题,本文在无监督密集匹配框架下,借鉴多个有监督网络结构,分别在航空影像数据集和作为参照的近景数据集上测试了匹配精度,实现了网络结构模块与精度关系的定性分析,为进一步探索深度学习在测绘领域的实用化提供了重要的参考。试验在相同损失函数条件下,分别采用DispNetS、DispNetC、iResNet、GCNet、PSMNetB及PSMNetS网络结构进行测试。经分析,得出如下结论：①测试的网络结构中,PSMNetS在航空影像数据集和近景数据集上表现稳定,且精度最高,训练整体耗时少,具有实用化的潜力;②在监督方法中效果更好的网络结构在无监督方法中效果不一定更好,其精度不仅取决于网络自身的匹配能力,同时也依赖于网络与损失函数的兼容性;③孪生网络模块、相关信息融合模块、金字塔池化模块和堆叠沙漏模块与无监督损失函数兼容性良好,可提升网络精度,而iResNet的图像重构迭代精化模块与重构损失函数重复约束,会产生“负优化”的作用。     

Grid pattern recognition in road networks using the C4.5 algorithm

Pattern recognition in road networks can be used for different applications, including spatiotemporal data mining, automated map generalization, data matching of different levels of detail, and other important research topics. Grid patterns are a common pattern type. This paper proposes and implements a method for grid pattern recognition based on the idea of mesh classification through a supervised learning process. To train the classifier, training datasets are selected from worldwide city samples with different cultural, historical, and geographical environments. Meshes are subsequently labeled as composing or noncomposing grids by participants in an experiment, and the mesh measures are defined while accounting for the mesh’s individual characteristics and spatial context. The classifier is generated using the C4.5 algorithm. The accuracy of the classifier is evaluated using Kappa statistics and the overall rate of correctness. The average Kappa value is approximately 0.74, which corresponds to a total accuracy of 87.5%. Additionally, the rationality of the classifier is evaluated in an interpretation step. Two other existing grid pattern recognition methods were also tested on the datasets, and comparison results indicate that our approach is effective in identifying grid patterns in road networks.     

基于深度学习的航空遥感影像密集匹配

本文探讨了深度学习在航空影像密集匹配中的性能,并与经典方法进行了比较,对模型泛化能力进行了评估。首先,实现了MC-CNN(matching cost convolutional neural network)、GC-Net(geometry and context network)、DispNet(disparity estimation network)3种代表性卷积神经元网络在航空立体像对上的训练和测试,并与传统方法SGM(semi-global matching)和商业软件SURE进行了比较。其次,利用直接迁移学习方法,评估了各模型在不同数据集间的泛化能力。最后,利用预训练模型和少量目标数据集样本,评估了模型微调的效果。试验包含3套航空影像、2套开源街景影像。试验表明:①与传统的遥感影像密集匹配方法相比,目前深度学习方法略有优势;②GC-Net与MC-CNN表现了良好的泛化能力,在开源数据集上训练的模型可以直接应用于遥感影像,且3PE(3-pixel-error)精度没有明显下降;③在训练样本不足时,利用预训练模型做初值并进行参数微调可以得到比直接训练更好的结果。     

基于语义类型本体的地名组配查询

地名查询方式多种多样,但它们都没有考虑地名语义类型因素,而语义类型恰恰是地名信息中相当重要的一部分。以地名语义分类为基础,构建了地名语义类型本体模型,并以郑州市为例进行了地名组配查询实验。结果表明,该方式对于检索地名的相关信息具有较好的辅助作用,为地名查询与检索作了有益的补充。     

A holistic approach to aligning geospatial data with multidimensional similarity measuring

Semantically aligning the heterogeneous geospatial datasets (GDs) produced by different organizations demands efficient similarity matching methods. However, the strategies employed to align the schema (concept and property) and instances are usually not reusable, and the effects of unbalanced information tend to be neglected in GD alignment. To solve this problem, a holistic approach is presented in this paper to integrally align the geospatial entities (concepts, properties and instances) simultaneously. Spatial, lexical, structural and extensional similarity metrics are designed and automatically aggregated by means of approval voting. The presented approach is validated with real geographical semantic webs, Geonames and OpenStreetMap. Compared with the well-known extensional-based aligning system, the presented approach not only considers more information involved in GD alignment, but also avoids the artificial parameter setting in metric aggregation. It reduces the dependency on specific information, and makes the alignment more robust under the unbalanced distribution of various information.     

An Automated Approach for Land Cover Classification Based on a Fuzzy Supervised Learning Framework

This paper proposes an automatic framework for land cover classification. In majority of published work by various researchers so far, most of the methods need manually mark the label of land cover types. In the proposed framework, all the information, like land cover types and their features, is defined as prior knowledge achieved from land use maps, topographic data, texture data, vegetation’s growth cycle and field data. The land cover classification is treated as an automatically supervised learning procedure, which can be divided into automatic sample selection and fuzzy supervised classification. Once a series of features were extracted from multi-source datasets, spectral matching method is used to determine the degrees of membership of auto-selected pixels, which indicates the probability of the pixel to be distinguished as a specific land cover type. In order to make full use of this probability, a fuzzy support vector machine (SVM) classification method is used to handle samples with membership degrees. This method is applied to Landsat Thematic Mapper (TM) data of two areas located in Northern China. The automatic classification results are compared with visual interpretation. Experimental results show that the proposed method classifies the remote sensing data with a competitive and stable accuracy, and demonstrate that an objective land cover classification result is achievable by combining several advanced machine learning methods.     

Ontology-driven discovering model for geographical information services

This paper proposes an ontology-driven discovering model for the geographical information services to improve their recall ratio and precision ratio. This model uses the geographical information service ontology. In this paper, first we study the multilevel matching arithmetic of geographical information services. This arithmetic is used for filtering and matching the services in the service register center according to the similarity between services selected and services requested from the definition of t...     

基于本体属性的基础地理信息概念语义相似性计算模型

语义相似性对于知识自动共享与集成起着非常重要的作用。在许多地理信息的应用领域,通常直接将分类体系作为领域(或任务)本体,并基于此计算概念间的语义距离以实现相似度计算。该方法虽然能够快速、简便地计算概念间的语义相似度,但是有时却因为分类体系的改变而造成相同概念间的相似度产生差异,甚至可能是错误的计算结果。本文面向基础地理信息领域,利用属性枚举方法表达概念的本质语义特征,从基础地理信息概念的内涵出发,提出基于本体属性的语义相似性计算模型。该模型将每个概念表达为本体属性集合,利用相关本体属性的相似性,结合权重信息计算概念的相似性。最后从基础地理信息概念中提取出100组样本,计算概念间的语义相似度并验证基于本体属性模型的有效性。实验结果表明基于本体属性的模型能更合理地计算出基础地理信息概念的相似度。     

基于Sentinel-1A数据的多种机器学习算法识别冰山的比较

冰山识别对于海洋环境监测和船只安全运行等具有重要的意义,是北极航道开通和北极开发过程中的重要内容。采用合成孔径雷达(SAR)影像进行冰山识别具有独特的优势,多种机器学习算法均可用于SAR影像的冰山识别中。为了最大限度地发挥机器学习算法的性能,有必要对不同机器学习算法及其搭配使用的特征与特征标准化方法进行评估,从而进行最优冰山识别方法的选择。因此,本文基于Sentinel-1A SAR影像,采用多种机器学习方法、多种特征组合及多种特征标准化方法进行冰山识别,并比较各流程方法的识别性能差异。采用的机器学习算法包括贝叶斯分类器(Bayes)、反向神经网络(BPNN)、线性判别分析(LDA)、随机森林(RF)以及支持向量机(SVM);特征标准化方法包括Min-max标准化、Z-score标准化及log函数标准化;数据集是含有12个SAR影像特征的969个冰山与非冰山样本,样本主要位于格陵兰岛东海岸。分类效果采用接收者操作特性(ROC)曲线下的面积(AUC)进行衡量。结果显示,最佳搭配下的RF的AUC值最高,达到了0.945,比最差的Bayes高出0.09。从识别率上来看,RF在冰山查全率为80%的情况下非冰山查全率达到92.6%,效果最好,比第2位的BPNN高出1.4%,比最差的Bayes高出2.6%;BPNN在冰山查全率为90%的情况下非冰山查全率达到87.4%,比第2位的RF高出0.8%,比最差的Bayes高出2.7%。上述结果表明,对冰山识别而言,选择最优的机器学习算法和最佳的特征与特征标准化方法都是十分重要的。     

Methods for matching English language addresses

Addresses occupy a niche location within the landscape of textual data, due to the positional importance carried by every word, and the geographic scope it refers to. The task of matching addresses happens every day and is present in various fields such as mail redirection, entity resolution, etc. Our work defines, and formalizes a framework to generate matching and mismatching pairs of addresses in the English language, and use it to evaluate various methods to automatically perform address matching. These methods vary widely from distance-based approaches to deep learning models. By studying the Precision, Recall, and Accuracy metrics of these approaches, we obtain an understanding of the best suited method for this setting of the address matching task.     

Multiple spectral similarity metrics for surface materials identification using hyperspectral data

Modern hyperspectral imaging and non-imaging spectroradiometer has the capability to acquire high-resolution spectral reflectance data required for surface materials identification and mapping. Spectral similarity metrics, due to their mathematical simplicity and insensitiveness to the number of reference labelled spectra, have been increasingly used for material mapping by labelling reflectance spectra in hyperspectral data labelling. For a particular hyperspectral data set, the accuracy of spectral labelling depends considerably upon the degree of unambiguous spectral matching achieved by the spectral similarity metric used. In this work, we propose a new methodology for quantifying spectral similarity for hyperspectral data labelling for surface materials identification. Developed adopting the multiple classifier system architecture, the proposed methodology unifies into a single framework the differential performances of eight different spectral similarity metrics for the quantification of spectral matching for surface materials. The proposed methodology has been implemented on two types of hyperspectral data viz. image (airborne hyperspectral images) and non-image (library spectra) for numerous surface materials identification. Further, the performance of the proposed methodology has been compared with the support vector machines (SVM) approach, and with all the base spectral similarity metrics. The results indicate that, for the hyperspectral images, the performance of the proposed methodology is comparable with that of the SVM. For the library spectra, the proposed methodology shows a consistently higher (increase of about 30% when compared to SVM) classification accuracy. The proposed methodology has the potential to serve as a general library search method for materials identification using hyperspectral data.     

Assessing the suitability of data from Sentinel-1A and 2A for crop classification

Sentinel-1A C-SAR and Sentinel-2A MultiSpectral Instrument (MSI) provide data applicable to the remote identification of crop type. In this study, six crop types (beans, beetroot, grass, maize, potato, and winter wheat) were identified using five C-SAR images and one MSI image acquired during the 2016 growing season. To assess the potential for accurate crop classification with existing supervised learning models, the four different approaches namely kernel-based extreme learning machine (KELM), multilayer feedforward neural networks, random forests, and support vector machine were compared. Algorithm hyperparameters were tuned using Bayesian optimization. Overall, KELM yielded the highest performance, achieving an overall classification accuracy of 96.8%. Evaluation of the sensitivity of classification models and relative importance of data types using data-based sensitivity analysis showed that the set of VV polarization data acquired on 24 July (Sentinel-1A) and band 4 data (Sentinel-2A) had the greatest potential for use in crop classification.