A Spatial Entropy-Based Decision Tree for Classification of Geographical Information

A decision tree is a classification algorithm that automatically derives a hierarchy of partition rules with respect to a target attribute of a large dataset. However, spatial autocorrelation makes conventional decision trees underperform for geographical datasets as the spatial distribution is not taken into account. The research presented in this paper introduces the concept of a spatial decision tree based on a spatial diversity coefficient that measures the spatial entropy of a geo‐referenced dataset. The principle of this solution is to take into account the spatial autocorrelation phenomena in the classification process, within a notion of spatial entropy that extends the conventional notion of entropy. Such a spatial entropy‐based decision tree integrates the spatial autocorrelation component and generates a classification process adapted to geographical data. A case study oriented to the classification of an agriculture dataset in China illustrates the potential of the proposed approach.     

Using Topological Relationships to Inform a Data Integration Process

When spatial datasets are overlaid, corresponding features do not always coincide. This may be a result of the datasets having differing quality characteristics, being captured at different scales or perhaps being in different projections or datums. Data integration methods have been developed to bring such datasets into alignment. Although these methods attempt to maintain topological relationships within each dataset, spatial relationships between features in different datasets are generally not considered. The preservation of inter‐dataset topology is a research area of considerable current interest. This research addresses the preservation of topology within a data integration process. It describes the functional models established to represent a number of spatial relationships as observation equations. These are used to provide additional information concerning the relative positions of features. Since many topological relationships are best modelled as inequalities, an algorithm is developed to accommodate such relationships. The method, based on least squares with inequalities (LSI), is tested on simulated and real datasets. Results are presented to illustrate the optimal positioning solutions determined using all of the available information. In addition, updated quality parameters are provided at the level of the individual coordinate, enabling communication of local variation in the resultant quality of the integrated datasets.     

Complexity reduction in choropleth map animations by autocorrelation weighted generalization of time-series data

Choropleth map animation is widely used to show the development of spatial processes over time. Although animation congruently depicts change, the rapid succession of complex map scenes easily exceeds the human cognitive capacity, causing map users to miss important information. Hence, a reduction of the visual complexity of map animations is desirable. This article builds on research related to complexity reduction of static choropleth maps. It proposes value generalization of choropleth time-series data in space and time, by using a method that adapts to the degree of global spatiotemporal autocorrelation within the dataset. A combination with upstream algorithms for local outlier detection results in less complex map animations focusing on large-scale patterns while still preserving significant local deviations in space and time. An according software application allows for in-depth exploration of the spatial and temporal autocorrelation structures in time-series data and provides control over the whole process of generalization.     

Evaluating Local Non-Stationarity when Considering the Spatial Variation of Large-scale Autocorrelation

Multi‐scale effects of spatial autocorrelation may be present in datasets. Given the importance of detecting local non‐stationarity in many theoretical as well as applied studies, it is necessary to “remove” the impact of large‐scale autocorrelation before common techniques for local pattern analysis are applied. It is proposed in this paper to employ the regionalized range to define spatially varying sub‐regions within which the impact of large‐scale autocorrelation is minimized and the local patterns can be investigated. A case study is conducted on crime data to detect crime hot spots and cold spots in San Antonio, Texas. The results confirm the necessity of treating the non‐stationarity of large‐scale spatial autocorrelation prior to any action aiming at detecting local autocorrelation.     

Assessing reference dataset representativeness through confidence metrics based on information density

Land cover maps obtained from classification of remotely sensed imagery provide valuable information in numerous environmental monitoring and modeling tasks. However, many uncertainties and errors can directly or indirectly affect the quality of derived maps. This work focuses on one key aspect of the supervised classification process of remotely sensed imagery: the quality of the reference dataset used to develop a classifier. More specifically, the representative power of the reference dataset is assessed by contrasting it with the full dataset (e.g. entire image) needing classification. Our method is applicable in several ways: training or testing datasets (extracted from the reference dataset) can be compared with the full dataset. The proposed method moves beyond spatial sampling schemes (e.g. grid, cluster) and operates in the multidimensional feature space (e.g. spectral bands) and uses spatial statistics to compare information density of data to be classified with data used in the reference process. The working hypothesis is that higher information density, not in general but with respect to the entire classified image, expresses higher confidence in obtained results. Presented experiments establish a close link between confidence metrics and classification accuracy for a variety of image classifiers namely maximum likelihood, decision tree, Backpropagation Neural Network and Support Vector Machine. A sensitivity analysis demonstrates that spatially-continuous reference datasets (e.g. a square window) have the potential to provide similar classification confidence as typically-used spatially-random datasets. This is an important finding considering the higher acquisition costs for randomly distributed datasets. Furthermore, the method produces confidence maps that allow spatially-explicit comparison of confidence metrics within a given image for identification of over- and under-represented image portions. The current method is presented for individual image classification but, with sufficient evaluation from the remote sensing community it has the potential to become a standard for reference dataset reporting and thus allowing users to assess representativeness of reference datasets in a consistent manner across different classification tasks.     

GPS sidereal filtering: coordinate- and carrier-phase-level strategies

Multipath error is considered one of the major errors affecting GPS observations. One can benefit from the repetition of satellite geometry approximately every sidereal day, and apply filtering to help minimize this error. For GPS data at 1 s interval processed using a double-difference strategy, using the day-to-day coordinate or carrier-phase residual autocorrelation determined with a 10-h window leads to the steadiest estimates of the error-repeat lag, although a window as short as 2 h can produce an acceptable value with > 97% of the optimal lag’s correlation. We conclude that although the lag may vary with time, such variation is marginal and there is little advantage in using a satellite-specific or other time-varying lag in double-difference processing. We filter the GPS data either by stacking a number of days of processed coordinate residuals using the optimum “sidereal” lag (23 h 55 m 54 s), and removing these stacked residuals from the day in question (coordinate space), or by a similar method using double-difference carrier-phase residuals (observational space). Either method results in more consistent and homogeneous set of coordinates throughout the dataset compared with unfiltered processing. Coordinate stacking reduces geometry-related repeating errors (mainly multipath) better than carrier-phase residual stacking, although the latter takes less processing time to achieve final filtered coordinates. Thus, the optimal stacking method will depend on whether coordinate precision or computational time is the over-riding criterion.     

A Space-Time GIS Approach to Exploring Large Individual-based Spatiotemporal Datasets

The increasing number of large individual-based spatiotemporal datasets in various research fields has challenged the GIS community to develop analysis tools that can efficiently help researchers explore the datasets in order to uncover useful information. Rooted in Hägerstrand's time geography, this study presents a generalized space-time path (GSTP) approach to facilitating visualization and exploration of spatiotemporal changes among individuals in a large dataset. The fundamental idea of this approach is to derive a small number of representative space-time paths (i.e. GSTPs) from the raw dataset by identifying spatial cluster centers of observed individuals at different time periods and connecting them according to their temporal sequence. A space-time GIS environment is developed to implement the GSTP concept. Different methods of handling temporal data aggregation and the creation of GSTPs are discussed in this article. Using a large individual-based migration history dataset, this study successfully develops an operational space-time GIS prototype in ESRI's ArcScene and ArcMap to provide a proof-of-concept study of this approach. This space-time GIS system demonstrates that the proposed GSTP approach can provide a useful exploratory analysis and geovisualization environment to help researchers effectively search for hidden patterns and trends in such datasets.     

A framework for connecting two interoperability universes: OGC Web Feature Services and Linked Data

Diverse studies have shown that about 80% of all available data are related to a spatial location. Most of these geospatial data are available as structured and semi‐structured datasets, and often use distinct data models, are encoded using ad‐hoc vocabularies, and sometimes are being published in non‐standard formats. Hence, these data are isolated within silos and cannot be shared and integrated across organizations and communities. Spatial Data Infrastructures (SDIs) have emerged and contributed to significantly enhance data discovery and accessibility based on OGC (Open Geospatial Consortium) Web services. However, finding, accessing, and using data disseminated through SDIs are still difficult for non‐expert users. Overcoming the current geospatial data challenges involves adopting the best practices to expose, share, and integrate data on the Web, that is, Linked Data. In this article, we have developed a framework for generating, enriching, and exploiting geospatial Linked Data from multiple and heterogeneous geospatial data sources. This proposal allows connecting two interoperability universes (SDIs, more specifically Web Feature Services, WFS, and Semantic Web technologies), which is evaluated through a study case in the (geo)biodiversity domain.     

Identification of multi-scale corresponding object-set pairs between two polygon datasets with hierarchical co-clustering

In this paper, we propose a means of finding multi-scale corresponding object-set pairs between two polygon datasets by means of hierarchical co-clustering. This method converts the intersection-ratio-based similarities of two objects from two datasets, one from each dataset, into the objects’ proximity in a geometric space using a Laplacian-graph embedding technique. In this space, the method finds hierarchical object clusters by means of agglomerative hierarchical clustering and separates each cluster into object-set pairs according to the datasets to which the objects belong. These pairs are evaluated with a matching criterion to find geometrically corresponding object-set pairs. We applied the proposed method to the segmentation result of a composite image with 6 NDVI images and a forest inventory map. Regardless of the different origins of the datasets, the proposed method can find geometrically corresponding object-set pairs which represent hierarchical distinctive forest areas.     

Improving the Elevation Accuracy of Digital Elevation Models: A Comparison of Some Error Detection Procedures

The widespread availability of powerful desktop computers, easy‐to‐use software tools and geographic datasets have raised the quality problem of input data to be a crucial one. Even though accuracy has been a concern in every serious application, there are no general tools for its improvement. Some particular ones exist, however, and some results are presented here for a particular case of quantitative raster data: Digital Elevation Models (DEM). Two procedures designed to detect anomalous values (also named gross errors, outliers or blunders) in DEMs, but valid also for other quantitative raster datasets, were tested. A DEM with elevations varying from 181 to 1044 m derived from SPOT data has been used as a contaminated sample, while a manually derived DEM obtained from aerial photogrammetry was regarded as the ground truth to allow a direct performance comparison for the methods with real errors. It is assumed that a “better” value can be measured or obtained through some methodology once an outlier location is suggested. The options are different depending upon the user (DEM producers might go to the original data and make another reading, while end users might use interpolation). Both choices were considered in this experiment. Preliminary results show that for the available dataset, the accuracy might be improved to some extent with very little effort. Effort is defined here as the percentage of points suggested by the methodology in relation with its total number: thus 100 per cent effort implies that all points have been checked. The method proposed by López (1997) gave poor results, because it has been designed for errors with low spatial autocorrelation (which is not the case here). A modified version was then designed and compared with the method proposed by Felicísimo (1994). The three procedures can be applied both for error detection during DEM generation and by end users, and they might be of use for other quantitative raster data. The choice of the best methodology is different depending on the effort involved. The conclusions have been derived for a photogrammetrically obtained DEM; other production procedures might lead to different results.     

The Influence of Land Cover on Shuttle Radar Topography Mission (SRTM) Elevations in Low‐relief Areas

The Shuttle Radar Topography Mission (SRTM), the first relatively high spatial resolution near‐global digital elevation dataset, possesses great utility for a wide array of environmental applications worldwide. This article concerns the accuracy of SRTM in low‐relief areas with heterogeneous vegetation cover. Three questions were addressed about low‐relief SRTM topographic representation: to what extent are errors spatially autocorrelated, and how should this influence sample design? Is spatial resolution or production method more important for explaining elevation differences? How dominant is the association of vegetation cover with SRTM elevation error? Two low‐relief sites in Louisiana, USA, were analyzed to determine the nature and impact of SRTM error in such areas. Light detection and ranging (LiDAR) data were employed as reference, and SRTM elevations were contrasted with the US National Elevation Dataset (NED). Spatial autocorrelation of errors persisted hundreds of meters spatially in low‐relief topography; production method was more critical than spatial resolution, and elevation error due to vegetation canopy effects could actually dominate the SRTM representation of the landscape. Indeed, low‐lying, forested, riparian areas may be represented as substantially higher than surrounding agricultural areas, leading to an inverted terrain model.     

Linear feature conflation: An optimization-based matching model with connectivity constraints

Geospatial data conflation is the process of combining multiple datasets about a geographic phenomenon to produce a single, richer dataset. It has received increased research attention due to its many applications in map making, transportation, planning, and temporal geospatial analyses, among many others. One approach to conflation, attempted from the outset in the literature, is the use of optimization-based conflation methods. Conflation is treated as a natural optimization problem of minimizing the total number of discrepancies while finding corresponding features from two datasets. Optimization-based conflation has several advantages over traditional methods including conciseness, being able to find an optimal solution, and ease of implementation. However, current optimization-based conflation methods are also limited. A main shortcoming with current optimized conflation models (and other traditional methods as well) is that they are often too weak and cannot utilize the spatial context in each dataset while matching corresponding features. In particular, current optimal conflation models match a feature to targets independently from other features and therefore treat each GIS dataset as a collection of unrelated elements, reminiscent of the spaghetti GIS data model. Important contextual information such as the connectivity between adjacent elements (such as roads) is neglected during the matching. Consequently, such models may produce topologically inconsistent results. In this article, we address this issue by introducing new optimization-based conflation models with structural constraints to preserve the connectivity and contiguity relation among features. The model is implemented using integer linear programming and compared with traditional spaghetti-style models on multiple test datasets. Experimental results show that the new element connectivity (ec-bimatching) model reduces false matches and consistently outperforms traditional models.     

How Reliable are Citizen‐Derived Scientific Data? Assessing the Quality of Contrail Observations Made by the General Public

Citizen science projects encourage the general public to participate in scientific research. Participants can contribute large volumes of data over broad spatial and temporal frames; however, the challenge is to generate data of sufficient quality to be useable in scientific research. Most observations made by citizen‐scientists can be independently verified by “experts”. However, verification is more problematic when the phenomena being recorded are short‐lived. This article uses a GIS methodology to verify the quality of contrail observations made by the general public as part of the OPAL Climate Survey. We verify observations using datasets derived from a variety of different sources (experts, models and amateur enthusiasts) with different spatial and temporal properties which reflect the complex 3D nature of the atmosphere. Our results suggest that ～70% of citizen observations are plausible, based on favorable atmospheric conditions and the presence or absence of aircraft; a finding which is in keeping with other, more conventional citizen science projects. However, questions remain as to why the quality of the citizen‐based observations was so high. Given the lack of supporting data on observers, it is impossible to determine whether the dataset was generated by the activities of many participants or a small but dedicated number of individual observers.     

顾及时空异质性的缺失数据时空插值方法

时空插值方法被广泛应用于缺失时空数据集的插值与估计。时空插值是时空建模与分析的一个重要内容,当前该研究关注的热点之一是异质条件下的时空插值与估计问题。因此,本文从时空数据的异质性出发,提出了一种顾及时空异质性的缺失数据时空插值方法。该方法首先对数据集进行时空分区,然后分别在时间和空间按照异质协方差模型计算缺失数据的估计值,进而利用相关系数确定时空权重、融合时间和空间估计值得到缺失数据的最终估计结果。最后通过两组气象数据集进行交叉验证对比分析试验。试验结果表明本文方法对比其他插值方法具有更高的精度和适用性。     

现实与赛博空间数据相结合的城市活动事件时空建模

城市活动事件(如文化、娱乐、体育等事件)的规模与影响力是城市经济文化发展的重要体现,其发生的全过程对城市现实空间与赛博空间都会产生巨大影响,从现实空间与赛博空间对城市活动事件的演化感知、动态建模与时空分析,具有重要的理论研究与应用价值。提出了一种结合现实空间交通数据与赛博空间社交媒体数据的城市活动事件时空建模分析方法,从事件进行中的交通轨迹,探测识别与事件显著相关的城市时空区域和交通流,分析现实空间事件热度的时空变化;从事件发生全过程的社交媒体数据中,探测分析赛博空间事件热度的时空变化;通过将现实空间和赛博空间的融合,建立城市活动事件时空模型,刻画事件全过程中城市地理空间与城市行为空间的时空演变特征。以2015年周杰伦"魔天伦2.0"世界巡回演唱会(武汉站)事件为例,采用武汉市出租车GPS轨迹数据和微博数据,对演唱会的事前、事中、事后实现城市地理空间与行为空间全过程建模与时空演变分析,并与单一数据源事件刻画模型进行比较,结果显示本方法能更合理地结合现实空间和赛博空间刻画城市活动事件。     

Spatially constrained regionalization with multilayer perceptron

In this article, multilayer perceptron (MLP) network models with spatial constraints are proposed for regionalization of geostatistical point data based on multivariate homogeneity measures. The study focuses on non‐stationarity and autocorrelation in spatial data. Supervised MLP machine learning algorithms with spatial constraints have been implemented and tested on a point dataset. MLP spatially weighted classification models and an MLP contiguity‐constrained classification model are developed to conduct spatially constrained regionalization. The proposed methods have been tested with an attribute‐rich point dataset of geological surveys in Ukraine. The experiments show that consideration of the spatial effects, such as the use of spatial attributes and their respective whitening, improve the output of regionalization. It is also shown that spatial sorting used to preserve spatial contiguity leads to improved regionalization performance.     

小数据集中的小型建筑物提取方法研究

遥感卫星影像一般尺寸较大，而包含有小型建筑物的区域占比较小，如果采用滑动切块扩增数据样本的方法，大部分切片中没有目标，而构建包含大量小建筑物的大型数据集费工费时。常规的方法在高分辨率卫星影像上提取小型建筑物非常困难，研究适用于小规模数据集的小型建筑物提取任务的提取方法具有重要理论意义和应用价值。本文提出了一种轻量化的全连接分割网络ZF-FCN，使用较小的感受野获取更多局部信息，使用较少的最大池化操作避免剧烈的下采样，使用Lovász-Softmax损失解决样本不平衡问题，使网络训练更稳定也更好地优化交并比。最后构建了一个主要包含小型建筑物的小规模数据集，试验在对不同切块大小进行数据增强后进行。对比试验表明，ZF-FCN在建筑物提取任务上的表现优于FCN和U-Net。     

Integrated compression of vehicle spatio‐temporal trajectories under the road stroke network constraint

With fast growth of all kinds of trajectory datasets, how to effectively manage the trajectory data of moving objects has received a lot of attention. This study proposes a spatio‐temporal data integrated compression method of vehicle trajectories based on stroke paths coding compression under the road stroke network constraint. The road stroke network is first constructed according to the principle of continuous coherence in Gestalt psychology, and then two types of Huffman tree—a road strokes Huffman tree and a stroke paths Huffman tree—are built, based respectively on the importance function of road strokes and vehicle visiting frequency of stroke paths. After the vehicle trajectories are map matched to the spatial paths in the road network, the Huffman codes of the road strokes and stroke paths are used to compress the trajectory spatial paths. An opening window algorithm is used to simplify the trajectory temporal data depicted on a time–distance polyline by setting the maximum allowable speed difference as the threshold. Through analysis of the relative spatio‐temporal relationship between the preceding and latter feature tracking points, the spatio‐temporal data of the feature tracking points are all converted to binary codes together, accordingly achieving integrated compression of trajectory spatio‐temporal data. A series of comparative experiments between the proposed method and representative state‐of‐the‐art methods are carried out on a real massive taxi trajectory dataset from five aspects, and the experimental results indicate that our method has the highest compression ratio. Meanwhile, this method also has favorable performance in other aspects: compression and decompression time overhead, storage space overhead, and historical dataset training time overhead.     

Towards semantic‐aware multiple‐aspect trajectory similarity measuring

The large amount of semantically rich mobility data becoming available in the era of big data has led to a need for new trajectory similarity measures. In the context of multiple‐aspect trajectories, where mobility data are enriched with several semantic dimensions, current state‐of‐the‐art approaches present some limitations concerning the relationships between attributes and their semantics. Existing works are either too strict, requiring a match on all attributes, or too flexible, considering all attributes as independent. In this article we propose MUITAS, a novel similarity measure for a new type of trajectory data with heterogeneous semantic dimensions, which takes into account the semantic relationship between attributes, thus filling the gap of the current trajectory similarity methods. We evaluate MUITAS over two real datasets of multiple‐aspect social media and GPS trajectories. With precision at recall and clustering techniques, we show that MUITAS is the most robust measure for multiple‐aspect trajectories.     

A Hybrid Geocoding Methodology for Spatio‐Temporal Data

As tools for collecting data continue to evolve and improve, the information available for research is expanding rapidly. Increasingly, this information is of a spatio‐temporal nature, which enables tracking of phenomena through both space and time. Despite the increasing availability of spatio‐temporal data, however, the methods for processing and analyzing these data are lacking. Existing geocoding techniques are no exception. Geocoding enables the geographic location of people and events to be known and tracked. However, geocoded information is highly generalized and subject to various interpolation errors. In addition, geocoding for spatio‐temporal data is especially challenging because of the inherent dynamism of associated data. This article presents a methodology for geocoding spatio‐temporal data in ArcGIS that utilizes several additional supporting procedures to enhance spatial accuracy, including the use of supplementary land use information, aerial photographs and local knowledge. This hybrid methodology allows for the tracking of phenomenon through space and over time. It is also able to account for reporting inconsistencies, which is a common feature of spatio‐temporal data. The utility of this methodology is demonstrated using an application to spatio‐temporal address records for a highly mobile group of convicted felons in Hamilton County, Ohio.