Table of Contents

Semantic segmentation on high-resolution aerial images plays a significant role in many remote sensing applications. Although the Deep Convolutional Neural Network (DCNN) has shown great performance in this task, it still faces the following two challenges: intra-class heterogeneity and inter-class homogeneity. To overcome these two problems, a novel dual-path DCNN, which contains a spatial path and an edge path, is proposed for high-resolution aerial image segmentation. The spatial path, which combines the multi-level and global context features to encode the local and global information, is used to address the intra-class heterogeneity challenge. For inter-class homogeneity problem, a Holistically-nested Edge Detection (HED)-like edge path is employed to detect the semantic boundaries for the guidance of feature learning. Furthermore, we improve the computational efficiency of the network by employing the backbone of MobileNetV2. We enhance the performance of MobileNetV2 with two modifications: (1) replacing the standard convolution in the last four Bottleneck Residual Blocks (BRBs) with atrous convolution; and (2) removing the convolution stride of 2 in the first layer of BRBs 4 and 6. Experimental results on the ISPRS Vaihingen and Potsdam 2D labeling dataset show that the proposed DCNN achieved real-time inference speed on a single GPU card with better performance, compared with the state-of-the-art baselines. Full article

Oriented feature from the accelerated segment test (oFAST) and rotated binary robust independent elementary features (rBRIEF) SLAM2 (ORB-SLAM2) represent a recognized complete visual simultaneous location and mapping (SLAM) framework with visual odometry as one of its core components. Given the accumulated error problem with RGB-Depth ORB-SLAM2 visual odometry, which causes a loss of camera tracking and trajectory drift, we created and implemented an improved visual odometry method to optimize the cumulative error. First, this paper proposes an adaptive threshold oFAST algorithm to extract feature points from images and rBRIEF is used to describe the feature points. Then, the fast library for approximate nearest neighbors strategy is used for image rough matching, the results of which are optimized by progressive sample consensus. The image matching precision is further improved by using an epipolar line constraint based on the essential matrix. Finally, the efficient Perspective-n-Point method is used to estimate the camera pose and a least-squares optimization problem is constructed to adjust the estimated value to obtain the final camera pose. The experimental results show that the proposed method has better robustness, higher image matching accuracy and more accurate determination of the camera motion trajectory. Full article

Frequent and rapid spatially explicit assessment of socioeconomic development is critical for achieving the Sustainable Development Goals (SDGs) at both national and global levels. Over the past decades, scientists have proposed many methods for estimating human activity on the Earth’s surface at various spatiotemporal scales using Defense Meteorological Satellite Program Operational Line System (DMSP-OLS) nighttime light (NTL) data. However, the DMSP-OLS NTL data and the associated processing methods have limited their reliability and applicability for systematic measuring and mapping of socioeconomic development. This study utilized Visible Infrared Imaging Radiometer Suite (VIIRS) NTL and the Isolation Forest machine learning algorithm for more intelligent data processing to capture human activities. We used machine learning and NTL data to map gross domestic product (GDP) at 1 km². We then used these data products to derive inequality indexes (e.g., Gini coefficients) at nationally aggregate levels. This flexible approach processes the data in an unsupervised manner at various spatial scales. Our assessments show that this method produces accurate subnational GDP data products for mapping and monitoring human development uniformly across the globe. Full article

A comprehensive fire risk assessment is very important in dense urban areas as it provides an estimation of people at risk and property. Fire policy and mitigation strategies in developing countries are constrained by inadequate information, which is mainly due to a lack of capacity and resources for data collection, analysis, and modeling. In this research, we calculated the fire risk considering two aspects, urban infrastructure and the characteristics of a high-rise building for a dense urban area in Zanjan city. Since the resources for this purpose were rather limited, a variety of information was gathered and information fusion techniques were conducted by employing spatial analyses to produce fire risk maps. For this purpose, the spatial information produced using unmanned aerial vehicles (UAVs) and then attribute data (about 150 characteristics of each high-rise building) were gathered for each building. Finally, considering high-risk urban infrastructures, like the position of oil and gas pipes and electricity lines and the fire safety analysis of high-rise buildings, the vulnerability map for the area was prepared. The fire risk of each building was assessed and its risk level was identified. Results can help decision-makers, urban planners, emergency managers, and community organizations to plan for providing facilities and minimizing fire hazards and solve some related problems to reduce the fire risk. Moreover, the results of sensitivity analysis (SA) indicate that the social training factor is the most effective causative factor in the fire risk. Full article

Natural hazards have a great number of influencing factors. Machine-learning approaches have been employed to understand the individual and joint relations of these factors. However, it is a challenging process for a machine learning algorithm to learn the relations of a large parameter space. In this circumstance, the success of the model is highly dependent on the applied parameter reduction procedure. As a state-of-the-art neural network model, representative learning assumes full responsibility of learning from feature extraction to prediction. In this study, a representative learning technique, recurrent neural network (RNN), was applied to a natural hazard problem. To that end, it aimed to assess the landslide problem by two objectives: Landslide susceptibility and inventory. Regarding the first objective, an empirical study was performed to explore the most convenient parameter set. In landslide inventory studies, the capability of the implemented RNN on predicting the subsequent landslides based on the events before a certain time was investigated respecting the resulting parameter set of the first objective. To evaluate the behavior of implemented neural models, receiver operating characteristic analysis was performed. Precision, recall, f-measure, and accuracy values were additionally measured by changing the classification threshold. Here, it was proposed that recall metric be utilized for an evaluation of landslide mapping. Results showed that the implemented RNN achieves a high estimation capability for landslide susceptibility. By increasing the network complexity, the model started to predict the exact label of the corresponding landslide initiation point instead of estimating the susceptibility level. Full article

Building information modeling (BIM) has received significant research attention in the field of built heritage. As-built BIM refers to a BIM representation of the “as-is” conditions of built heritage at the time of a survey. Determining the level of development (LoD) is crucial for as-built BIM owing to its relevance to model effects and modeling efforts. This study addresses this issue from the viewpoint of a brick structure based on a case study of a fifteenth-century ruin in Nanjing, China. Three LoDs are proposed based on the combined use of a commercial platform and auxiliary tools: A host model linked with raster images composed using orthoimage and relief maps (LoD 1), an as-built volume with semantic skins (LoD 2), and a brick-by-brick model with custom industry foundation class parameters at local areas (LoD 3). The results reveal that LoD 1 caters to an efficient web-based workflow for brick-damage annotations; as-built dimensions can be extracted from LoD 2; and LoD 3 enables attributes, such as damage types, to be attached at the brick level. In future studies, the detection of brick shapes is expected to automate the process of as-built surface mapping. Full article

Virtual 3D city models can be stored and exchanged in the CityGML open data model. When dynamic phenomena in 3D cities are represented with a CityGML application domain extension (ADE), the objects in CityGML are often used as static background, and it is difficult to represent the evolutionary process of the objects themselves. Although a construction process model in building information modeling (BIM) is available, it cannot efficiently and accurately simulate the building construction process at the city level. Accordingly, employing the arrow diagramming method, we developed a CityGML ADE to represent this process. We extended the hierarchy of the model and proposed the process levels of detail model. Subsequently, we explored a mechanism to associate the construction process and building objects as well as the mechanism to automate construction process transitions. Experiments indicated that the building construction process ADE (BCPADE) could adequately express the characteristics of this process. Compared with the building construction process model in the architecture, engineering, and construction field, BCPADE removes redundant information, i.e., that unrelated to a 3D city. It can adequately express building construction processes at multiple spatiotemporal scales and accurately convey building object behavior during building evolution, such as adding, removal, merging, and change. Such characteristics enable BCPADE to render efficient and accurate simulations of the building construction process at the city level. Full article

Social vulnerability, from a socio-environmental point of view, focuses on the identification of disadvantaged or vulnerable groups and the conditions and dynamics of the environments in which they live. To understand this issue, it is important to identify the factors that explain the difficulty of facing situations with a social disadvantage. Due to its complexity and multidimensionality, it is not always easy to point out the social groups and urban areas affected. This research aimed to assess the connection between certain dimensions of social vulnerability and its urban and dwelling context as a fundamental framework in which it occurs using a decision model useful for the planning of social and urban actions. For this purpose, a holistic approximation was carried out on the census and demographic data commonly used in this type of study, proposing the construction of (i) a knowledge model based on Artificial Neural Networks (Self-Organizing Map), with which a demographic profile is identified and characterized whose indicators point to a presence of social vulnerability, and (ii) a predictive model of such a profile based on rules from dwelling variables constructed by conditional inference trees. These models, in combination with Geographic Information Systems, make a decision model feasible for the prediction of social vulnerability based on housing information. Full article

Many real-world dynamic features such as ocean eddies, rain clouds, and air masses may split or merge while they are migrating within a space. Topologically, the migration trajectories of such features are structurally more complex as they may have multiple branches due to the splitting and merging processes. Identifying the spatial aggregation patterns of the trajectories could help us better understand how such features evolve. We propose a method, a Global Similarity Measuring Algorithm for the Complex Trajectories (GSMCT), to examine the spatial proximity and topologic similarity among complex trajectories. The method first transforms the complex trajectories into graph structures with nodes and edges. The global similarity between two graph structures (i.e., two complex trajectories) is calculated by averaging their topologic similarity and the spatial proximity, which are calculated using the Comprehensive Structure Matching (CSM) and the Hausdorff distance (HD) methods, respectively. We applied the GSMCT, the HD, and the Dynamic Time Warping (DTW) methods to examine the complex trajectories of the 1993–2016 mesoscale eddies in the South China Sea (SCS). Based on the similarity evaluation results, we categorized the complex trajectories across the SCS into four groups, which are similar to the zoning results reported in previous studies, though difference exists. Moreover, the yearly numbers of complex trajectories in the clusters in the northernmost (Cluster 1) and the southernmost SCS (Cluster 4) are almost the same. However, their seasonal variation and migration characteristics are totally opposite. Such new knowledge is very useful for oceanographers of interest to study and numerically simulate the mesoscale ocean eddies in the SCS. Full article

Recently, an increasing number of cities have deployed bicycle-sharing systems to solve the first/last mile connection problem, generating a large quantity of data. In this paper, singular value decomposition (SVD) was used to extract the main features of the cycling flow from the origin and destination (OD) data of shared bicycles in Beijing. The results show that (1) pairs of OD flow clusters can be derived from the pairs of vectors after SVD, and each pair of clusters represents a small part of an area with dockless shared bicycles; (2) the spatial clusters derived from the top vectors of SVD are highly coincident with the hot spot areas in the heatmap of shared bicycles; (3) approximately 30% of the study area accounts for nearly 80% of bike riding; (4) nearly 70% of the clustered area derived from the top 1000 vectors of SVD is associated with subway stations; and (5) the types of point of interest (POI) differ between the origin area and destination area for the clustered area of the top 1000 vectors. Full article

Metropolitan cities are facing many socio-economic problems (e.g., frequent traffic congestion, unexpected emergency events, and even human-made disasters) related to urban crowd flows, which can be described in terms of the gathering process of a flock of moving objects (e.g., vehicles, pedestrians) towards specific destinations during a given time period via different travel routes. Understanding the spatio-temporal characteristics of urban crowd flows is therefore of critical importance to traffic management and public safety, yet it is very challenging as it is affected by many complex factors, including spatial dependencies, temporal dependencies, and environmental conditions. In this research, we propose a novel matrix-computation-based method for modeling the morphological evolutionary patterns of urban crowd flows. The proposed methodology consists of four connected steps: (1) defining urban crowd levels, (2) deriving urban crowd regions, (3) quantifying their morphological changes, and (4) delineating the morphological evolution patterns. The proposed methodology integrates urban crowd visualization, identification, and correlation into a unified and efficient analytical framework. We validated the proposed methodology under both synthetic and real-world data scenarios using taxi mobility data in Wuhan, China as an example. Results confirm that the proposed methodology can enable city planners, municipal managers, and other stakeholders to identify and understand the gathering process of urban crowd flows in an informative and intuitive manner. Limitations and further directions with regard to data representativeness, data sparseness, pattern sensitivity, and spatial constraint are also discussed. Full article

Road extraction is a unique and difficult problem in the field of semantic segmentation because roads have attributes such as slenderness, long span, complexity, and topological connectivity, etc. Therefore, we propose a novel road extraction network, abbreviated HsgNet, based on high-order spatial information global perception network using bilinear pooling. HsgNet, taking the efficient LinkNet as its basic architecture, embeds a Middle Block between the Encoder and Decoder. The Middle Block learns to preserve global-context semantic information, long-distance spatial information and relationships, and different feature channels’ information and dependencies. It is different from other road segmentation methods which lose spatial information, such as those using dilated convolution and multiscale feature fusion to record local-context semantic information. The Middle Block consists of three important steps: (1) forming a feature resource pool to gather high-order global spatial information; (2) selecting a feature weight distribution, enabling each pixel position to obtain complementary features according to its own needs; and (3) inversely mapping the intermediate output feature encoding to the size of the input image by expanding the number of channels of the intermediate output feature. We compared multiple road extraction methods on two open datasets, SpaceNet and DeepGlobe. The results show that compared to the efficient road extraction model D-LinkNet, our model has fewer parameters and better performance: we achieved higher mean intersection over union (71.1%), and the model parameters were reduced in number by about 1/4. Full article

Building information model (BIM) data are digital and geometric-based data that are enriched thematically, semantically, and relationally, and are conceptually very similar to geographic information. In this paper, we propose both the use of the international standard ISO 19157 for the adequate formulation of the quality control for BIM datasets and a statistical approach based on a binomial/multinomial or hypergeometric (univariate/multivariate) model and a multiple hypothesis testing method. The use of ISO 19157 means that the definition of data quality units conforms to data quality elements and well-defined scopes, but also that the evaluation method and conformity levels use standardized measures. To achieve an accept/reject decision for quality control, a statistical model is needed. Statistical methods allow one to limit the risks of the parties (producer and user risks). In this way, several statistical models, based on proportions, are proposed and we illustrate how to apply several quality controls together (multiple hypothesis testing). All use cases, where the comparison of a BIM dataset versus reality is needed, are appropriate situations in which to apply this method in order to supply a general digital model of reality. An example of its application is developed to control an “as-built” BIM dataset where sampling is needed. This example refers to a simple residential building with four floors, composed of a basement garage, two commercial premises, four apartments, and an attic. The example is composed of six quality controls that are considered simultaneously. The controls are defined in a rigorous manner using ISO 19157, by means of categories, scopes, data quality elements, quality measures, compliance levels, etc. The example results in the rejection of the BIM dataset. The presented method is, therefore, adequate for controlling BIM datasets Full article

The heavy industry in India has witnessed rapid development in the past decades. This has increased the pressures and load on the Indian environment, and has also had a great impact on the world economy. In this study, the Preparatory Project Visible Infrared Imaging Radiometer (NPP VIIRS) 375-m active fire product (VNP14IMG) and night-time light (NTL) data were used to study the spatiotemporal patterns of heavy industrial development in India. We employed an improved adaptive K-means algorithm to realize the spatial segmentation of long-term VNP14IMG data and artificial heat-source objects. Next, the initial heavy industry heat sources were distinguished from normal heat sources using a threshold recognition model. Finally, the maximum night-time light data were used to delineate the final heavy industry heat sources. The results suggest, that this modified method is a much more accurate and effective way of monitoring heavy industrial heat sources, and the accuracy of this detection model was higher than 92.7%. The number of main findings were concluded from the study: (1) the heavy industry heat sources are mainly concentrated in the north-east Assam state, east-central Jharkhand state, north Chhattisgarh and Odisha states, and the coastal areas of Gujarat and Maharashtra. Many heavy industrial heat sources were also found around a line from Kolkata on the Eastern Indian Ocean to Mumbai on the Western Indian Ocean. (2) The number of working heavy industry heat sources (NWH) and, particularly, the total number of fire hotspots for each working heavy industry heat source area (NFHWH) are continuing to increase in India. These trends mirror those for the Gross Domestic Product (GDP) and total population of India between 2012 and 2017. (3) The largest values of NWH and NFHWH were in Jharkhand, Chhattisgarh, and Odisha whereas the smallest negative values, the $Slope\_NWH$ in Jharkhand and Chhattisgarh were also the two largest values in the whole country. The smallest negative values of $Slope\_NWH$ and $Slope\_NFHWH$ were in Haryana. The $Slope\_NFHWH$ in the mainland Gujarat had the second most negative value, while the value of the $Slope\_NWH$ was the third-highest positive value. Full article

The discipline of the humanities has long been inseparable from the exploration of space and time. With the rapid advancement of digitization, databases, and data science, humanities research is making greater use of quantitative spatiotemporal analysis and visualization. In response to this trend, our team developed the Chinese academic map publishing platform (AMAP) with the aim of supporting the digital humanities from a Chinese perspective. In compiling materials mined from China’s historical records, AMAP attempts to reconstruct the geographical distribution of entities including people, activities, and events, using places to connect these historical objects through time. This project marks the beginning of the development of a comprehensive database and visualization system to support humanities scholarship in China, and aims to facilitate the accumulation of spatiotemporal datasets, support multi-faceted queries, and provide integrated visualization tools. The software itself is built on Harvard’s WorldMap codebase, with enhancements which include improved support for Asian projections, support for Chinese encodings, the ability to handle long text attributes, feature level search, and mobile application support. The goal of AMAP is to make Chinese historical data more accessible, while cultivating collaborative opensource software development. Full article

Traditional Geographic Information Systems (GIS) represent the environment under reductionist thinking, which disaggregates a geographic environment into independent geographic themes. The reductionist approach makes the spatiotemporal characteristics of geo-features explicit, but neglects the holistic nature of the environment, such as the hierarchical structure and interactions among environmental elements. To fill this gap, we integrate the concept geographic scenario with the fundamental principles of General System Theory to realize the environmental complexity in GIS. With the integration, a geographic scenario constitutes a hierarchy of spatiotemporal frameworks for organizing environmental elements and subserving the exploration of their relationships. Furthermore, we propose geo-characterization with ontological commitments to both static and dynamic properties of a geographic scenario and prescribe spatial, temporal, semantic, interactive, and causal relationships among environmental elements. We have tested the utility of the proposed representation in OWL and the associated reasoning process in Semantic Web Rule Language (SWRL) rules in a case study in Nanjing, China. The case study represents Nanjing and the Nanjing presidential palace to demonstrate the connections among environmental elements in different scenarios and the support for information queries, evolution process simulation, and semantic inferences. The proposed representation encodes geographic knowledge of the environment, makes the interactions among environmental elements explicit, supports geographic process simulation, opens opportunities for deep knowledge mining, and grounds a foundation for GeoAI to discover geographic complexity and dynamics beyond the support of conventional theme-centric inquiries in GIS. Full article

Roads represent a major source of mortality for many species. To mitigate road mortality, it is essential to know where collisions with vehicles are happening and which species and populations are most affected. For this, moving platforms such as mobile mapping systems (MMS) can be used to automatically detect road-killed animals on the road surface. We recently developed an MMS to detect road-killed amphibians, composed of a scanning system on a trailer. We present here a smaller and improved version of this system (MMS2) for detecting road-killed amphibians and small birds. It is composed of a stereo multi-spectral and high definition camera (ZED), a high-power processing laptop, a global positioning system (GPS) device, a support device, and a lighter charger. The MMS2 can be easily attached to any vehicle and the surveys can be performed by any person with or without sampling skills. To evaluate the system’s effectiveness, we performed several controlled and real surveys in the Évora district (Portugal). In real surveys, the system detected approximately 78% of the amphibians and birds present on surveyed roads (overlooking 22%) and generated approximately 17% of false positives. Our system can improve the implementation of conservation measures, saving time for researchers and transportation planning professionals. Full article

“BugMap” is a citizen science mobile application that provides a platform for amateur and expert scientists to report sightings of two invasive insect pests, the tiger mosquito Aedes albopictus Skuse (Diptera: Culicidae) and the brown marmorated stink bug, Halyomorpha halys Stål (Hemiptera: Pentatomidae). The latter is a notorious pest of fruit trees, vegetables, ornamentals, and row crops, inflicting severe agricultural and ecological disturbances in invaded areas. Our approach consists of coupling traditional monitoring with citizen science to uncover H. halys invasion in Trentino. The project was initiated in 2016 and the first results were reported in 2018. Here, we revisit our initiative four years after its adoption and unravel new information related to the invader dispersal and overwintering capacity. We found that our previous model predicted the current distribution of H. halys in Trentino with an accuracy of 72.5%. A new MaxEnt model was generated by pooling all reports received so far, providing a clearer perspective on areas at risk of stink bug establishment in this north Italian region. The information herein presented is of immediate importance for enhancing monitoring strategies of this pest and for refining its integrated management tactics. Full article

Analyses of the correlations between social and economic phenomena are rarely limited to simple evaluations of the relationships that exist between two features. Information about the structure and behaviour of complex phenomena and processes in the natural environment and social systems is usually incomplete and uncertain. Grey relational analysis (GRA) poses an alternative to statistical methods (e.g., correlation analysis, variance analysis, regression analysis and direct comparisons) to evaluate complex phenomena. In GRA, the number of assumptions relating to the size and distribution of samples is far smaller than in statistical methods. The required number of observations in the GRA is n ≥ 4. Therefore, the grey system theory (GST) provides useful tools for analysing limited and imperfect data. GST can be used to predict a system’s future behaviour and to evaluate the relationships between observation vectors. The study aimed to determine the strength of the relationships between the analysed features with the use of GST and to analyse the model’s behaviour for a different number of variables. The main assumptions and definitions relating to GST were presented. The residential preferences of a selected social group were analysed. The proposed approach supports the development of effective decision-making procedures in urban planning. Full article

For improved drought planning and response, there is an increasing need for highly predictive and stable drought prediction models. This paper presents the performance of both homogeneous and heterogeneous model ensembles in the satellite-based prediction of drought severity using artificial neural networks (ANN) and support vector regression (SVR). For each of the homogeneous and heterogeneous model ensembles, the study investigates the performance of three model ensembling approaches: (1) non-weighted linear averaging, (2) ranked weighted averaging, and (3) model stacking using artificial neural networks. Using the approach of “over-produce then select”, the study used 17 years of satellite data on 16 selected variables for predictive drought monitoring to build 244 individual ANN and SVR models from which 111 models were automatically selected for the building of the model ensembles. Model stacking is shown to realize models that are superior in performance in the prediction of future drought conditions as compared to the linear averaging and weighted averaging approaches. The best performance from the heterogeneous stacked model ensembles recorded an R² of 0.94 in the prediction of future (1 month ahead) vegetation conditions on unseen test data (2016–2017) as compared to an R² of 0.83 and R² of 0.78 for ANN and SVR, respectively, in the traditional approach of selection of the best (champion) model. We conclude that despite the computational resource intensiveness of the model ensembling approach, the returns in terms of model performance for drought prediction are worth the investment, especially in the context of the continued exponential increase in computational power and the potential benefits of improved forecasting for vulnerable populations. Full article

This work discusses the integration of multi-camera video moving objects (MCVO) and GIS. This integration was motivated by the characteristics of multi-camera videos distributed in the urban environment, namely, large data volume, sparse distribution and complex spatial–temporal correlation of MCVO, thereby resulting in low efficiency of manual browsing and retrieval of videos. To address the aforementioned drawbacks, on the basis of multi-camera video moving object extraction, this paper first analyzed the characteristics of different video-GIS Information fusion methods and investigated the integrated data organization of MCVO by constructing a spatial–temporal pipeline among different cameras. Then, the conceptual integration model of MCVO and GIS was proposed on the basis of spatial mapping, and the GIS-MCVO prototype system was constructed in this study. Finally, this study analyzed the applications and potential benefits of the GIS-MCVO system, including a GIS-based user interface on video moving object expression in the virtual geographic scene, video compression storage, blind zone trajectory deduction, retrieval of MCVO, and video synopsis. Examples have shown that the integration of MCVO and GIS can improve the efficiency of expressing video information, achieve the compression of video data, rapidly assisting the user in browsing video objects from multiple cameras. Full article

Human activity type inference has long been the focus for applications ranging from managing transportation demand to monitoring changes in land use patterns. Today’s ever increasing volume of mobility data allow researchers to explore a wide range of methodological approaches for this task. Such data, however, lack reference observations that would allow the validation of methodological approaches. This research proposes a methodological framework for urban activity type inference using a Dirichlet multinomial dynamic Bayesian network with an empirical Bayes prior that can be applied to mobility data of low spatiotemporal resolution. The method was validated using open source Foursquare data under different isochrone configurations. The results provide evidence of the limits of activity detection accuracy using such data as determined by the Area Under Receiving Operating Curve (AUROC), log-loss, and accuracy metrics. At the same time, results demonstrate that a hierarchical modeling framework can provide some flexibility against the challenges related to the nature of unsupervised activity classification using trajectory variables and POIs as input. Full article

Accurate information on the number and distribution of pedestrians in space and time helps urban planners maintain current city infrastructure and design better public spaces for local residents and visitors. Previous studies have demonstrated that using webcams together with crowdsourcing platforms to locate pedestrians in the captured images is a promising technique for analyzing pedestrian activity. However, it is challenging to efficiently transform the time series of pedestrian locations in the images to information suitable for geospatial analytics, as well as visualize data in a meaningful way to inform urban design or decision making. In this study, we propose to use a space-time cube (STC) representation of pedestrian data to analyze the spatio-temporal patterns of pedestrians in public spaces. We take advantage of AMOS (The Archive of Many Outdoor Scenes), a large database of images captured by thousands of publicly available, outdoor webcams. We developed a method to obtain georeferenced spatio-temporal data from webcams and to transform them into high-resolution continuous representation of pedestrian densities by combining bivariate kernel density estimation with trivariate, spatio-temporal spline interpolation. We demonstrate our method on two case studies analyzing pedestrian activity of two city plazas. The first case study explores daily and weekly spatio-temporal patterns of pedestrian activity while the second one highlights the differences in pattern before and after plaza’s redevelopment. While STC has already been used to visualize urban dynamics, this is the first study analyzing the evolution of pedestrian density based on crowdsourced time series of pedestrian occurrences captured by webcam images. Full article

This exploratory study identifies spatial patterns of crimes and their associations with the index of Unsatisfied Basic Needs (UBN), with Communitarian Policy Units (CPU) density, as well as with population density. The case study is the Metropolitan District of Quito. Correlation analyses were applied between number of registers of each type of crime, and the UBN index, CPU density and population density measures. The spatial autocorrelation index of Getis-Ord Gi* was calculated to identify hotspots of the different types of crime. Ordinary least squares regressions and geographically weighted regressions considering types of crime as dependent variables, were calculated. Larceny and robbery were found to be the predominant crimes in the study area. An inverse relationship between the UBN index and number of crimes was identified for each type of crime, while positive relationships were found between crimes and CPU density, and between crimes and population density. Significant hotspots of fraud, homicide, larceny, murder, rape and robbery were found in all urban parishes. Additionally, crime hotspots were identified in eastern rural parishes adjacent to urban parishes. This study provides important implications for crime prevention in the Metropolitan District of Quito (MDQ), and the obtained results contribute to the ecology of crime research in the study area. Full article

High-precision land use/land cover classification mapping derived from remote sensing supplies essential datasets for scientific research on environmental assessment, climate change simulation, geographic condition monitoring, and environmental management at global and regional scales. It is an important issue in the study of earth system science, and the coastal area is a hot spot region in this field. In this paper, the coastal areas of the Maritime Silk Road were used as the research object and a fusion method based on agreement analysis and fuzzy-set theory was adopted to achieve the fusion of three land use/land cover datasets: MCD12Q1-2010, CCI-LC2010, and GlobeLand30-2010. The accuracy of the fusion results was analyzed using an error matrix, spatial confusion, average overall consistency, and average type-specific consistency. The main findings were as follows. (1) After the establishment of reference data based on Google Earth, both the producer accuracy and user accuracy of the fusion data were improved when compared with those of the three input data sources, and the fusion data had the highest overall accuracy and Kappa coefficient, with values of 90.37% and 0.8617, respectively. (2) Various input data sources differed in terms of the correctly classified contributions and misclassified influences of different land use/land cover types in the fusion data; furthermore, the overall accuracy and Kappa coefficient between the fusion data and any one of the input data sources were far higher than those between any two of the input data sources. (3) The average overall consistency of the fusion data was the highest at 89.29%, which was approximately 5% higher than that of the input data sources. (4) The average type-specific consistencies of cropland, forest, grassland, shrubland, wetland, artificial surfaces, bare land, and permanent snow and ice in the fusion data were the highest, with values of 69.95%, 74.41%, 21.24%, 34.22%, 97.62%, 51.83%, 84.39%, and 2.46%, respectively; compared with the input data sources, the average type-specific consistencies of the fusion data were 0.61–20.32% higher. This paper provides information and suggestions for the development and accuracy evaluation of future land use/land cover data in global and regional coastal areas. Full article

Technological advances have led to numerous developments in data sources. Geo-tagged photo metadata has provided a new source of mass research data for tourism studies. A series of data processing methods centering on the various types of information contained in geo-tagged photo metadata have thus been proposed; as a result, the development of tourism studies based on such data has advanced. However, an in-depth study of the data processing methods designed to conduct tourist flow prediction based on geo-tagged photo metadata has not yet been conducted. In order to acquire accurate substitutive data regarding inbound flows in cities, this paper introduces and designs several methods, including data screening, text data similarity calculation, geographical location clustering, and time series data modelling, in order to realize a data preprocessing model for inbound tourist flows in cities based on geo-tagged photo metadata. Wherein, the entropy filtering method was introduced to aid in determining whether the data were posted by inbound tourists; whether the inbound persons’ activities were related to tourism was judged through the calculation of tag text similarity; an efficient clustering method based on geographic grid partition was designed for cases in which the tag values were empty; finally, the time series of the inbound tourist flows of a certain region and period were obtained through data statistics and normalization. For the empirical research, Beijing City in China was selected as the research case, after which the feasibility and accuracy of the methods proposed in this paper were verified through data correlation analysis between Flickr data and real statistical yearbook data, as well as analysis of the prediction results based on a machine learning algorithm. The data preprocessing method introduced and designed in this paper provides a reference for the study of geo-tagged photo metadata in the field of tourism flow prediction. These methods can effectively filter out inbound tourist flow data from geotag photo metadata, thus providing a novel, reliable, and low-cost research data source for urban inbound tourism flow forecasting. Full article

Urban areas involve different functions that attract individuals and fit personal needs. Understanding the distribution and combination of these functions in a specific district is significant for urban development in cities. Many researchers have already studied the methods of identifying the dominant functions in a district. However, the degree of collection and the representativeness of a function in a district are controlled not only by its number in the district but also by the number outside this district and a number of other functions. Thus, this study proposed a quantitative method to identify urban functions, using Fisher’s exact test and point of interest (POI) data, applied in determining the urban districts within the Sixth Ring Road in Beijing. To begin with, we defined a functional score based on three statistical features: the p-value, odds-ratio, and the frequency of each POI tag. The p-value and odds-ratio resulted from a statistical significance test, the Fisher’s exact test. Next, we ran a k-modes clustering algorithm to classify all urban districts in accordance with the score of each function and their combination in one district, and then we detected four different groups, namely, Work and Tourism Mixed-developed district, Mixed-developed Residential district, Developing Greenland district, and Mixed Recreation district. Compared with the other identifying methods, our method had good performance in identifying functions, except for transportation. In addition, the Coincidence Degree was used to evaluate the accuracy of classification. In our study, the total accuracy of identifying urban districts was 83.7%. Overall, the proposed identifying method provides an additional method to the various methods used to identify functions. Additionally, analyzing urban spatial structure can be simpler, which has certain theoretical and practical value for urban geospatial planning. Full article

The low-cost, indoor-feasibility, and non-intrusive characteristic of passive infrared sensors (PIR sensors) makes it widely used in human motion detection, but the limitation of its object identification ability makes it difficult to further analyze in the field of Geographic Information System (GIS). We present a template matching approach based on geometric algebra (GA) that can recover the semantics of different human motion patterns through the binary activation data of PIR sensor networks. A 5-neighborhood model was first designed to represent the azimuth of the sensor network and establish the motion template generation method based on GA coding. Full sets of 36 human motion templates were generated and then classified into eight categories. According to human behavior characteristics, we combined the sub-sequences of activation data to generate all possible semantic sequences by using a matrix-free searching strategy with a spatiotemporal constraint window. The sub-sequences were used to perform the matching operation with the generation-templates. Experiments were conducted using Mitsubishi Electric Research Laboratories (MERL) motion datasets. The results suggest that the sequences of human motion patterns could be efficiently extracted in different observation periods. The extracted sequences of human motion patterns agreed well with the event logs under various circumstances. The verification based on the environment and architectural space shows that the accuracy of the result of our method was up to 96.75%. Full article

Remotely sensed data can reinforce the abilities of water resources researchers and decision-makers to monitor water quality more effectively. In the past few decades, remote sensing techniques have been widely used to measure qualitative water quality parameters. However, the use of moderate resolution sensors may not meet the requirements for monitoring small water bodies. Water quality in a small dam was assessed using high-resolution satellite data from RapidEye and in situ measurements collected a few days apart. The satellite carries a five-band multispectral optical imager with a ground sampling distance of 5 m at its nadir and a swath width of 80 km. Several different algorithms were evaluated using Pearson correlation coefficients for electrical conductivity (EC), total dissolved soils (TDS), water transparency, water turbidity, depth, suspended particular matter (SPM), and chlorophyll-a. The results indicate strong correlation between the investigated parameters and RapidEye reflectance, especially in the red and red-edge portion with highest correlation between red-edge band and water turbidity (r² = 0.92). Two of the investigated indices showed good correlation in almost all of the water quality parameters with correlation higher than 0.80. The findings of this study emphasize the use of both high-resolution remote sensing imagery and red-edge portion of the electromagnetic spectrum for monitoring several water quality parameters in small water areas. Full article