GISTAM 2023 Abstracts

Area 1 - Data Acquisition and Processing
Full Papers Short Papers
Area 2 - Remote Sensing
Full Papers Short Papers
Area 3 - Modeling, Representation and Visualization
Full Papers Short Papers
Area 4 - Knowledge Extraction and Management
Short Papers
Area 5 - Domain Applications
Full Papers Short Papers

Area 1 - Data Acquisition and Processing

Full Papers
Paper Nr: 8
Title:

Spatio-Temporal Modelling of Relationship Between Organic Carbon Content and Land Use Using Deep Learning Approach and Several Co-Variables: Application to the Soils of the Beni Mellal in Morocco
Authors:

Sébastien Gadal, Mounir Oukhattar, Catherine Keller and Ismaguil H. Houmma
Abstract: In recent decades, population growth has led to rapid urbanisation associated with a land degradation process that threatens soil organic carbon stocks (SOCS). This paper aims to model the interrelationships between SOCS and land use/land cover (LULC). The approach was based on the use of environmental covariates derived from Landsat-5 TM/8 OLI images, forty soil samples, Kriging spatial interpolation method and a Multi-layer Perceptron (MLP) model for the geo-spatialisation of SOCS. The analysis shows a high positive autocorrelations (R2>0.75) between vegetation indices and SOCS, particularly higher for SOCS derived from spatial modelling with MLP. On the other hand, the relationship between LULC and SOCS from the three approaches is very variable depending on the dynamics of LULC. The autocorrelations between SOCS and LULC units are very weak in 1985 and 2000 but significant for the year 2018. This suggests that the land use dynamics in the area are favourable to SOCS. In general, the results show that SOCS increased in the tree crop, unused land and forest areas but decreased in the cropland. The SOCS varied in the following order: forest cover>unused land>cropland>urban area>tree crops. This indicates that LULC, topography and vegetation types had an impact on SOCS distribution characteristics.
Download
Short Papers
Paper Nr: 7
Title:

Revisiting Food Deserts in North Carolina, USA, Using a Cloud-Based Real-Time Quality Assurance/Quality Control (QA/QC) Tool
Authors:

Timothy Mulrooney and Isabel Gutierrez
Abstract: In the study of the food environment, little research has explored the spatial data quality of store locations which impacts the spatial representation of the food environment. In this paper, we created a cloud-based tool that can inspect, edit and create new supermarkets in real-time which changes the complexion of the food environment. Comparisons were made between data supplied between a CAB (Commercially Available Business) Database and those corrected after field verification. Results showed differences between the food environment using the data provided and the actual food environment after QA/QC, with a general underestimation of those who are truly food needy due to errors of temporal accuracy, misattribution and geocoding in the original data provided.
Download
Paper Nr: 12
Title:

LiDAR and SfM-MVS Integrated Approach to Build a Highly Detailed 3D Virtual Model of Urban Areas
Authors:

Nives Grasso, Claudio Spadavecchia, Vincenzo Di Pietra and Elena Belcore
Abstract: The three-dimensional reconstruction of buildings, road infrastructures, service networks, and cultural heritage in urban environments is relevant for many market segments and numerous functions in the management and coordination of public authorities. These stakeholders are showing increasing interest in modern acquisition and reconstruction technologies for digital models typical of the geomatic and computer vision disciplines. In this context, it is essential to methodically exploit the potential of active and passive instruments and apply multi-sensor integration techniques, to obtain metrically accurate and high-resolution products. This study proposes a multi-sensor and multi-scale approach for high-resolution 3D model reconstruction focused on a city portion of Turin (Italy). We performed an integrated survey based on LiDAR and photogrammetric techniques, both aerial and terrestrial. Then we produced a set of 3D digital products for (i) promoting the historical and artistic heritage through Virtual Reality (VR) applications, (ii) supporting the restoration of Baroque buildings, and (iii) providing advanced analysis concerning the alteration of the urban road system. The final output describes in detail the architectural elements investigated (e.g., 9,480,000 tringles to define the mesh of a statue). It emphasizes the need for deepening sensor integration and data fusion.
Download
Paper Nr: 19
Title:

GNSS Monitoring of Geodynamics in the Region Around Sofia and South-Western Bulgaria
Authors:

Nikolay Dimitrov and Anton Ivanov
Abstract: For more than 25 years, the monitoring of geodynamic processes with modern GNSS technology in the region of Sofia and Southwestern Bulgaria continues. To investigate modern crustal motions in the area, Global Positioning System (GPS) data obtained between 1996 and 2022 are analyzed to obtain the velocity field for southwestern Bulgaria. For this period, monitoring covered 28 stations. The active strain in the region is also estimated from analysis of the results for velocity solutions. Some correlation between modern earth crust movements, seismic events and tectonic structures is established. The obtained results in a general way confirm previously data, but with much better accuracy and details at local level. The results can be used for a detailed geodynamic and geological study of the active fault structures in the area.
Download
Paper Nr: 22
Title:

Standards-Based Geospatial Services Integration for Smart Cities Platforms
Authors:

Bruno Rienzi, Raquel Sosa, Gastón Abellá, Ana Machado, Daniel Susviela and Laura González
Abstract: Smart cities usually refer to the use of information and communication technologies to provide citizens with improved city services and quality of life, in an affordable and sustainable way. Geospatial technologies, especially those based on standards, are relevant to this purpose, as location is crucial for organising, processing, and analysing urban information and services. During the last years, many smart cities platforms have emerged to provide support for the design, implementation, deployment, and management of smart cities applications (e.g. FIWARE). Although these platforms frequently consider the spatial dimension, they do not usually provide native support for typical geospatial services (e.g. data access, portrayal, and processing services). Therefore, geospatial-related features provided by applications are usually developed from scratch and on a per-case basis, which leads to code duplication and hinders their implementation agility, maintainability, and reuse. This paper proposes a standards-based solution for geospatial services integration for smart cities platforms, which comprises an overall architecture as well as a reference implementation based on FIWARE and includes three smart cities applications.
Download
Paper Nr: 23
Title:

On-the-Fly Acquisition and Rendering with Low Cost LiDAR and RGB Cameras for Marine Navigation
Authors:

Somnath Dutta, Fabio Ganovelli and Paolo Cignoni
Abstract: This paper describes a hardware/software system, dubbed NausicaaVR, for acquiring and rendering 3D environments in the context of marine navigation. Like other similar work, it focuses on system calibration and rendering but the specific context poses new and more difficult challenges for the development when compared to the classic automotive scenario. We provide a comprehensive description of all the components of the system, explicitly reporting on encountered problems and subtle choices to overcome those, in an attempt to render an insightful picture of how this and similar systems are built.
Download
Paper Nr: 24
Title:

Accuracy Assessment of Direct Georeferencing Using UAV Matrice 210 RTK V2 on Gully Santiš, Island of Pag (Croatia)
Authors:

Katarina Glavačević, Ivan Marić and Ante Šiljeg
Abstract: Rapid development and increased availability of unmanned aerial vehicles (UAVs) resulted in the exponential use of these systems in many scientific fields and activities. However, the application of photogrammetric models derived using the Structure from Motion (SfM) technique largely depends on the use of ground control points (GCPs). Since the acquisition of the GCPs requires the use of high-quality total stations or GNSS-RTK receivers, these procedures generally take up a lot of time. Execution of a photogrammetric process without using the GCPs is called direct georeferencing, and it is becoming an increasingly popular method. In this research, we tested three methods of RTK positioning using the system of the Matrice 210 RTK V2 and D-RTK 2 mobile station. The following methods were tested: (a) D-RTK 2 as a base station; (b) D-RTK 2 correction with the third-party base station; (c) network NTRIP corrections CROPOS. An absolute accuracy assessment of each RTK positioning mode was done using 10 check points (CPs). By calculating the total RMSE, it was determined that (b) and (c) RTK positioning modes have a centimeter level of accuracy (<10 cm). In this research, it is determined that the tested UAV system for direct georeferencing can be used in a wide range of geographical applications and other disciplines where absolute accuracy of centimeter-level is required.
Download
Paper Nr: 27
Title:

New Challenges in the Implementation and Exploitation of a Low-Cost Web Map of the Active Deformation Areas Across Europe
Authors:

José A. Navarro, Danielly García and Michele Crosetto
Abstract: The European Ground Motion Service (EGMS) is already offering data on Persistent Scatterers (PS) throughout Europe, which will aid in analyzing ground deformation on a continental scale. However, to more fully comprehend ground motion processes, it is preferable to use Active Deformation Areas (ADA) instead of PS. The CTTC has been using its ADAfinder tool to generate ADAs since 2018. With the new availability of EGMS PS data for the entire continent, the CTTC is now working on producing ADAs for all of Europe and making them accessible to the public through a self-developed, in-house hosted, web-based map application. A former paper describes the initial steps taken to develop it. This work focuses on how the challenge of processing a huge amount of data has affected the design and implementation of the tools used in the data production workflow. Additionally, the manner in which the EGMS data is structured, providing data sets that spatially overlap, has resulted in a new problem: overlapping ADAs. The approach to resolving this issue is also discussed in this paper. The result is an evolution of the initial concept where not only economic reasons but also considerations on automation and handling of large volumes of data have guided the design and implementation of the system.
Download
Paper Nr: 9
Title:

Bushfire Susceptibility Mapping Using Gene Expression Programming and Machine Learning Methods: A Case Study of Kangaroo Island, South Australia
Authors:

Maryamsadat Hosseini and Samsung Lim
Abstract: Kangaroo Island, South Australia is one of the bushfire-prone areas. A catastrophic bushfire known as the black summer hit Kangaroo Island in 2019/2020. We chose Kangaroo Island as a case study to generate bushfire susceptibility maps using five different methods, namely gene expression programming (GEP), random forest (RF), support vector machine (SVM), frequency ratio (FR) and logistic regression (LR). To generate bushfire susceptibility maps, we used eight contributing factors including: digital elevation model, slope, aspect, normalized difference vegetation index, distance to roads, distance to streams, precipitation, and land cover. The proposed methods were evaluated by area under the curves (AUCs) of receiver operating characteristic. RF performed best with an AUC of 0.93, followed by SVM and GEP with AUCs equal to 0.89 and 0.88, respectively, but LR and FR performed least among the five methods with AUCs 0.85 and 0.84, respectively. The generated bushfire susceptibility maps show that western and central areas of Kangaroo Island are highly vulnerable to bushfire.
Download
Paper Nr: 30
Title:

Biodiversity, Urban Quality Life and Air Quality Indices for Hotspot Detection of Transformation Opportunities in Cities: A Case Study in Barcelona
Authors:

Danielly García, Mariona Ferrandiz-Rovira, Oriol Serra and M. E. Parés
Abstract: Half of the world’s population lives in cities, where usually there are few little green space and there are also high levels of air pollution. Moreover, the traditional urbanization of cities contributes to climate change, promotes the loss of global biodiversity and induces serious health problems for citizens. Both climate change and the loss of biodiversity affect negatively to the ecosystems and therefore human health, as they are responsible for providing clean air, food, fresh water, medicines, renewable resources... This deterioration increases significantly the risk of human-borne infectious diseases such as coronavirus or HIV. The ability we have to re-naturalize anthropogenic spaces and learn to generate spaces for coexistence will be key for the future of our society. The research presented in this paper aims to do a step forward to achieve that ability by working in three schools of the city of Barcelona and their surroundings. Among other actions, in this project, a diagnosis of neighborhood has been carried out. The diagnosis includes the identification and quantification of relevant indicators regarding neighborhood’s biodiversity and also the quality of daily life and the analysis of pollutants (NO2 and PM10 ) near the schools during the 2021-2022 school year. All these information has been merged in a single geographic data base and relevant hotspots where to act have been identified. The information has been shared with city council and citizens.
Download
Paper Nr: 38
Title:

GIS Multicriteria Decision Analysis in Selecting the Optimal Location for Urban Green Space: A Case Study of Zadar City
Authors:

Rina Milošević, Silvija Šiljeg and Fran Domazetović
Abstract: The urbanization process has proceeded rapidly in recent decades, resulting in the rapid transformation of natural surfaces into impervious ones which has numerous impacts on the environment and human health. Urban green spaces are recognized as a critical spatial component for maintaining ecological balance and improving human mental and physical health. Therefore, the rational and even distribution of green spaces in the city is particularly important, as they represent the most accessible natural environment for city dwellers. The main objective of this study is to propose criteria and create a UGS suitability model (UGSSM) for the urban area of Zadar. The model is generated by applying the GIS multicriteria decision analysis (MCDA) and analytical hierarchical process (AHP). The model resulted in 580 ha of very high suitable (VHS) zones, mostly located in the northwestern part of the city. However, only 0.05% (N=38) of VHS zones are consolidated areas larger than 2 ha. Among VHS consolidated areas (>2 ha), the optimal one is depicted based on ownership verification. This framework can be applied to other small cities with some minor modifications. For future research, we suggest including residents with physical disabilities in the selection and landscaping of the location.
Download
Paper Nr: 40
Title:

Deep-Learning Based Super-Resolution of Aeolianite Images on the Purpose of Edge Detection and Pattern Extraction
Authors:

Antigoni Panagiotopoulou, Lemonia Ragia and Niki Evelpidou
Abstract: In the current work processing of Aeolianite images, from a quarry in the island of Naxos in Greece, is presented. The deep-learning based technique called Densely Residual Laplacian Super-Resolution (DRLN) is applied on the original images of size 3000×4000 pixels to increase their spatial resolution per the factor of 4. Edge detection is applied on the initial images as well as on the super-resolved images of 12000×16000 pixels. Visual and numerical comparisons on several Aeolianite scenes prove that the super-resolved images are advantageous in relation to the initial images of lower spatial resolution, as far as edge detection and pattern delineation are concerned. The improvement in edge detected components reaches 83%. Classification or pattern extraction could significantly benefit from encompassing the proposed methodology for Aeolianite images as a preprocessing step.
Download

Area 2 - Remote Sensing

Full Papers
Paper Nr: 10
Title:

Sentinel 2 High-Resolution Land Cover Mapping in Sub-Saharan Africa with Google Earth Engine
Authors:

Elena Belcore and Marco Piras
Abstract: This work aims to develop an efficient methodology for high-resolution spatial and thematic land cover maps of sub-Saharan areas based on Sentinel-2 data. LC mapping in these areas is complicated due to their land morphology, climatic conditions and homogeneity of surface spectral responses. Two pixel-based supervised classification approaches are compared in Google Earth Engine. The aggregated method classifies each image and then aggregates the results on frequency bases at pixel level. The stacked method classifies all the images together in a single stacked database. Additionally, the influence of linear atmospheric correction models on the overall accuracy (OA) is assessed, and the best-performing approach is compared to existing Land Cover (LC) maps of the area. 16 Sentinel-2 images (level 1C) from 2017 and 2019 were atmospheric and topographically corrected and classified into nine classes. The results show similar performances for the analysed approaches, with a slightly high OA for the aggregated classification (0.97). The atmospheric correction has little impact on the results.
Download
Paper Nr: 17
Title:

A Novel Method for InSAR Phase Unwrapping with Single Baseline
Authors:

Chenxi Tian and Guoman Huang
Abstract: The precision of phase unwrapping (PU), one of the primary methods used in interferometric synthetic aperture radar (InSAR), has a direct impact on the accuracy of the digital elevation model (DEM) that InSAR produces. The phase continuity assumption restricts single-baseline (SB) PU, and it is frequently hard to achieve optimal PU results in complex terrain areas with significant gradient variations. Fortunately, by utilizing numerous InSAR interferograms, or the elevation changes corresponding to each interference fringe in the interferogram, multi-baseline (MB) PU can totally overcome the restriction of the phase continuity assumption. Therefore, this paper proposes a virtual-baseline (VB) PU based on the two-stage programming (TSPA) MB PU approach to transform the SB PU problem into a MB PU problem. The novel method can be referred to as VB-TSPA. First, the effect of baseline length on MB PU is comprehensively considered to determine the virtual baseline length. Then a corresponding interferogram is simulated according to the length. Finally, the TSPA method is used for the MB PU. The experimental results from simulated and real data demonstrate that the novel PU method has a better effect than the traditional SB PU algorithm and can obtain higher precision DEM.
Download
Paper Nr: 36
Title:

Exploring Spectral Data, Change Detection Information and Trajectories for Land Cover Monitoring over a Fire-Prone Area of Portugal
Authors:

André Alves, Daniel Moraes, Bruno Barbosa, Hugo Costa, Francisco D. Moreira, Pedro Benevides, Mário Caetano and Manuel Campagnolo
Abstract: Land use/land cover (LULC) change detection and classification in maps based on automated data processing are becoming increasingly sophisticated in Earth Observation (EO). There is a growing number of annual maps available, with diverse but related production structures consisting primarily of classification and post-classification phases, the latter of which deals with inaccuracies of the first. The methodology production of the “Carta de Ocupação do Solo conjuntural” (COSc), a thematic land cover map of continental Portugal produced by the Directorate-General for Territory (DGT) mostly based on Sentinel-2 images classification, includes a semi-automatic phase of correction that combines expert knowledge and ancillary data in if-then-else rules validated by photointerpretation. Although this approach reduces misclassifications from an initial Random Forest (RF) prediction map, improving consistency between years and compliance with ecological succession, requires a lot of time-consuming semi-automatic procedures. This work evaluates the relevance of exploring an additional set of variables for automatic classification over disturbance-prone areas. A multitemporal dataset with 124 variables was analysed using data dimensionality reduction techniques, resulting in the identification of 35 major explanatory indicators, which were then used as inputs for RF classification with cross-validation. The estimated importance of the explanatory variables shows that composites of spectral bands, which are already included in the current COSc workflow, in conjunction with the inclusion of additional data namely, historical land cover information and change detection coefficients, from the Continuous Change Detection and Classification (CCDC) algorithm, are relevant for predicting land cover classes after disturbance. Since map updating is a more challenging task for disturbed pixels, we focused our analysis on locations where COSc indicated potential land cover change. Nonetheless, the overall classification accuracy for our experiments was 72.34 % which is similar to the accuracy of COSc for this region of Portugal. The findings suggest new variables that could improve future COSc maps.
Download
Paper Nr: 37
Title:

Identification of Emergent and Floating Aquatic Vegetation Using an Unsupervised Thresholding Approach: A Case Study of the Dniester Delta in Ukraine
Authors:

Ioannis Manakos, Eleftherios Katsikis, Sergiy Medinets, Yevgen Gazyetov, Leonidas Alagialoglou and Volodymyr Medinets
Abstract: Monitoring of emergent and floating vegetation in freshwater ecosystems is of high importance for water management in an area. This study proposes a methodology for the automatic monitoring of aquatic vegetation using indicators estimated via remote sensing image analysis. The study area is located in the Lower Dniester Basin in Southern Ukraine. The approach is developed using Sentinel-2 images and validated with field measurements. The goal is to discriminate and map three classes of aquatic surface condition; namely, areas covered with floating vegetation, or dominated by emergent vegetation, and open water. The approach is transferable across different dates over a period of three years. Results are useful for governmental authorities and natural/ national park administrations for near real-time monitoring of aquatic vegetation to mitigate the impact of overgrowth on water quality, biodiversity, and ecosystem services.
Download
Paper Nr: 39
Title:

Data-Driven Modelling of Freshwater Ecosystems: A Multiscale Framework Based on Global Geospatial Data
Authors:

Bruna Almeida and Pedro Cabral
Abstract: Freshwater ecosystems are primarily impacted by climate, land use and land cover changes, and over-abstraction. Satellite Earth observation (SEO) data and technologies are key in environmental modelling and support decisions. These technologies combined with machine learning (ML) are a powerful approach for modelling freshwater ecosystems at a multiscale level. The goal of this study is to present a set of reference data and guidelines that can be used to estimate the water and wetness probability index (WWPI) in different spatial and temporal scales. To find the best model’s predictors, sensitivity analyses were carried out in a predictive ML model implemented in a transnational river basin district (Portugal – Spain), the Tagus Basin. Satellite imagery, satellite-derived data, biophysical variables, and landscape characteristics were the explanatory variables evaluated in the sensitivity analyses, and some of them were included in the framework as a reference source of spatial data.
Download
Short Papers
Paper Nr: 31
Title:

Using Deep Learning and Radar Backscatter for Mapping River Water Surface
Authors:

Diana Orlandi, Federico A. Galatolo, Mario A. Cimino, Carolina Pagli, Nicola Perilli, Joao A. Pompeu and Itxaso Ruiz
Abstract: In the last decades, the effects of global warming combined with growing anthropogenic activities have caused a mismatch in the water supply-demand, resulting in a negative impact on numerous Mediterranean rivers regime and on the functionality of related ecosystem services. Thus, for water management and mitigation of the potential hazards, it is fundamental to efficiently map areal extents of river water surface. Synthetic Aperture Radar (SAR) is one of the satellite technologies applied for hydrological studies, but it has a spatial resolution which is limited for the study of rivers. On the other side, deep learning technology exhibits a high modelling potential with low spatial resolution data. In this paper, a method based on convolutional neural networks is applied to the SAR backscatter coefficient for detecting river water surface. Our experimental study focuses on the lower reach of Mijares river (Eastern Spain), covering a period from Apr 2019 to Sept 2022. Results suggest that radar backscattering has high potential in modelling water river trends, contributing to the monitoring of the effects of climate change and impacts on related ecosystem services. To assess the effectiveness of the method, the output has been validated with the Normalized Difference Water Index (NDWI).
Download
Paper Nr: 32
Title:

Imperviousness Density Mapping Based on GIS-MCDA and High-Resolution Worldview-2 Imagery
Authors:

Lovre Panđa
Abstract: Accurate monitoring and extraction of impervious surfaces are essential for urban planning and sustainable environmental management. Increasing urbanization has led to a significant increase in the extent of impervious surfaces, which, along with climate change, are the leading cause of increasingly frequent flooding in urban areas. To prevent flooding disasters in urban areas, flood hazard and risk analyses must be carried out. An imperviousness density model is one of the most important criteria in such analyses. In this study, an imperviousness density model of the city of Zadar was created using GIS-MCDA and four criteria (LULC, NDVI, slope and TWI). The criteria were extracted from WorldView-2 (WV-2) imagery and linearly standardized using the Fuzzy logic approach. The Analytic Hierarchy Process (AHP) was used to determine the final model for imperviousness density. The model with a spatial resolution of 0.5 m, based on the WV-2 imagery turned out to be much more detailed than existing publicly available models, such as the Copernicus imperviousness density model, which is based on Sentinel-2 imagery with a spatial resolution of 10 m.
Download

Area 3 - Modeling, Representation and Visualization

Full Papers
Paper Nr: 18
Title:

Borehole Inner Surface Visualization System with Vibration Cancellation and Trajectory Smoothing Based on Optical Monocular Video Camera
Authors:

Nan Zong, Waleed Al-Nuaimy, Heba Lakany and Paul Worthington
Abstract: The rapid digitization and modelling of the planet brings with it increased demand for the tools necessary to process and visualize disparate streams of multivariate and often highly complex geophysical data streams. The rationalization of detection hardware and the integration of sensors offers the potential to economically visually explore and map geophysical environments such as the interior of subterranean boreholes. This paper addresses the challenge of visual reconstruction of the geometry of the inner surface of a borehole from video data collected via a monocular optical camera. We introduce a novel system of algorithms to unwrap the cylindrical borehole inner surface data and to compensate for the offsets and errors arising during data acquisition. Three modules are designed for this task: Unwrapping module consisting o algorithms to generate visualization results of borehole inner surfaces; Vibration cancellation module that compensates for rotation and drift errors caused by the movement of detectors, balancing computational cost and performance; Trajectory smoothing based on image convolution signal processing methods to filter out anomalies and interruptions that arise as a result of the other processing stages. The proposed system integrates these modules to generate planar side-view images with a high level of spatial accuracy. This system also contributes to establish a novel and easy-to-access visualization tool of boreholes with simplified detectors that only consists of a monocular camera and a fixed circular LED band. Results has demonstrated the system is capable of resisting high frequency drift and the effects of rotation and vibrations in harsh subterranean environments. This novel combination of video and image processing marks a significant improvement over currently available or published borehole video exploration techniques, and can be further extended and enhanced to deliver more accurate multi-sensor 3D modeling and reconstruction of the complex inner structure of geophysical boreholes.
Download
Paper Nr: 33
Title:

Spatial Patterns in Neurodegenerative Disease’s Hospitalizations in Portugal (2000-2016)
Authors:

Mariana Oliveira, Alberto Freitas, Ana C. Teodoro and Hernâni Gonçalves
Abstract: Neurodegenerative diseases, usually arising from the death of nervous system cells, are a rising concern in the worlds’ population increasing life expectancy. More precisely, the Portuguese population, along with that of other developed countries, is ageing at a fast rate. The understanding of such diseases’ patterns is of utmost relevance to help manage the burden it represents in the health system. In this retrospective study, we analysed over 500 thousand hospitalizations with discharges between 2000 and 2016. We computed age-standardized hospitalization rates for each neurodegenerative disease. The most prevalent disease in our sample was Dementia with 43.4% of cases, and the least prevalent was Basal with only 0.2% of cases. The spatial analysis shows that Santarém and Portalegre (neighbour) districts in central Portugal, have the highest rates. The increase in hospitalization rates over the study period is also clear when looking at the spatio-temporal analysis. Although limited by the usage of secondary health data, this study represents a background for other studies on the field of neurodegenerative diseases, presenting with relevant insight into the spatio-temporal patterns of each and every neurodegenerative disease in Portugal at the moment.
Download
Short Papers
Paper Nr: 13
Title:

Modelling Spatial Connectivity of Forest Harvest Areas: Exact and Heuristic Approaches
Authors:

Pete Bettinger
Abstract: A forest management planning process can involve the development of a tactical plan that illustrates for a land manager where to go and what to do within a specific period of time, acknowledging and satisfying all recognized management constraints. More often these days, forest management constraints address the size, timing, and placement of management activities. The optimization methods used to mathematically develop a forest plan, and to integrate spatial constraints into planning efforts are often referred to as exact and heuristic approaches. This paper describes how one might model spatial connectivity of forest harvest areas as constraints under both approaches, using two different representations of connectivity, the unit restriction model and the area restriction model. The heuristic approach to the latter has until now only been described using scientific notation. Here, we provide guidance for the programming logic.
Download
Paper Nr: 21
Title:

Evaluating the Fulfilment Rate of Charging Demand for Electric Vehicles Using Open-Source Data
Authors:

Hana Elattar, Ferdinand von Tüllenburg, Sebastian Wöllmann and Javier Valdes
Abstract: With the shift towards electric vehicles accelerating; we are working with open-source data to estimate to which degree existing charging infrastructure is fulfilling the demand created by electric vehicles. This paper is explaining how to create such a calculation by extracting data from large public areas in the city of Lindau (Bodensee), Germany as a showcase. With this data we aim to evaluate whether charging stations located in the premises of public and commercial buildings cover the demand of electric vehicles reaching the said buildings. This research is conducted as a first step of methodologies development that aims on the long term to create a tool that supports in the optimal placement of new charging stations. The methodology chosen is inspired by two main concepts: the first is the attractiveness factor concept used for the creation of travel models, while the second is the classification of charging stations based on location to determine their rate of occupancy. They are both used to cluster buildings and charging stations respectively to be able to determine the number of users in the area of study (AOS) compared to the overall number of electric vehicles reaching the destination in a given day. This paper takes the island in the centre of the city of Lindau (Bodensee) as its area under investigation and uses open-source data along with the appropriate assumptions as a base for its calculations.
Download
Paper Nr: 41
Title:

Vector Tile Geospatial Data Protection Using Quantization-Based Watermarking
Authors:

Danila Glazkov, Nikolay Chupshev and Victor Fedoseev
Abstract: The paper proposes a watermarking method for protecting geodata presented in the Mapbox Vector Tile (MVT) format against theft. MVT is an open format that is gaining popularity in web mapping services due to efficient storage and fast rendering. However, the vector nature of the format makes it an easy target for attackers who want to steal data and use in their services. The method proposed in this paper protects MVT data with a digital watermark based on re-quantization of point coordinates of object geometry. The method can be adjusted using a number of parameters that allow finding a balance between the robustness of the digital watermark to map distortions and the error introduced when embedding. A series of experiments performed showed the robustness of this method to several distortions: removal of some objects and layers, reduction in the number of points of existing objects, addition of new objects, controlled shift of points in the tile geometry. With a proper choice of the watermark parameters, even with a moderate level of each of the listed distortions, which does not lead to a loss of significance of the protected geodata, the method can reach 100% watermark extraction accuracy of all bits of the built-in watermark.
Download

Area 4 - Knowledge Extraction and Management

Short Papers
Paper Nr: 29
Title:

Evaluation of Urban Perception Using Only Image Segmentation Features
Authors:

Xinyi Li, Benjamin Beaucamp, Vincent Tourre, Thomas Leduc and Myriam Servières
Abstract: Deep learning has been used with the street-view imagery Place Pulse 2.0 to evaluate the perception of urban space along six perceptual dimensions: safe, lively, beautiful, wealthy, boring, and depressing. Traditional methods automatically extract feature representations from images through a convolutional neural network to yield prediction. However, the formers are computationally intensive and do not take a priori into account the semantic information from panoptic segmentation scene. In light of this, we propose that learning with semantic information could be close to full image analysis for the prediction of perceptual qualities. A lightweight solution is presented, which quickly predicts the sense of urban space from the implied highly compressed segmentation feature vectors of the street-view images via deep/machine learning models. Our solution achieves an average accuracy of about 62%, which is acceptable compared to the baseline result accuracy of 68%, and significantly reduces the complexity of the data and the computational effort.
Download

Area 5 - Domain Applications

Full Papers
Paper Nr: 11
Title:

The Evolution of the South-Eastern Baltic Sea Coastline Between 1988 and 2018 by Remote Sensing
Authors:

Sébastien Gadal and Thomas Gloaguen
Abstract: This article aims to define and explain the evolution of the coastline in Latvia, Lithuania, and Russia since the late 1980s. Coastal erosion is a critical issue for public authorities and is considered as one of the main environmental problems in the south-eastern Baltic region. The political, economic, and social changes associated with the collapse of the Soviet Union have created new pressures in recent decades in previously relatively undeveloped coastal regions. The geomorphology of the latter is the result of various natural morpho-dynamic processes: swells, tides, tectonic movements, etc. Landsat 4-5 TM, Landsat 8 OLI satellite images series between 1988 and 2018 are used to estimate the position of the coastline. The spatial accuracy of the shoreline automatic recognition based on the combination of minimum noise fraction and Laplacian convolution operators is compared with the manual methods of photo interpretation. The results showed a global change of –0.21 m/year with local and temporal disparities. It can be explained by a variety of natural and anthropogenic factors that disrupt the sedimentary stock and the hydrodynamic forces controlling coastal evolution.
Download
Paper Nr: 16
Title:

3D Geospatial Data Management Architecture for Common Operational Picture Functionalities in Emergency Response
Authors:

Iñaki Cejudo, Eider Irigoyen, Harbil Arregui and Estíbaliz Loyo
Abstract: In disaster management, the emergency response commanded from the Command and Control center can make the difference for a faster and safer outcome. Novel sensing tools provide new capabilities to monitor and evaluate what happens in real time with location information of first responders and casualties being a key resource. Therefore, Geographical Information Systems (GIS) are essential when representing the Common Operational Picture to have a complete understanding of the situation in the field, obtained through big amounts of real-time data coming from multiple sources, and therefore support decision making. Moreover, the 3-dimensional representation of the terrain and buildings enhance the classical 2D map representation. In this work, a detailed overview of the architecture components and functionalities developed for a data-driven emergency response 3D web GIS application is described. In addition, a quantitative evaluation of how the number of location records collected from various numbers of first responders impacts the performance of some geospatial tasks needed for an efficient visual representation is provided, depending on the data measurement frequency.
Download
Short Papers
Paper Nr: 20
Title:

When Should We Report the Traffic Jams of Today? A Case Study on a Swiss Highway Using Graph Neural Networks and Expert Knowledge
Authors:

Jhonny Pincay, Ana Oña and Damian Nomura
Abstract: This case study manuscript details the conception and implementation of an artifact that uses floating car data to forecast average speeds on a segment of a Swiss national road. To consider the spatial and temporal dependencies when performing the predictions, the studied segment was modeled as a graph and as a time series problem. Subsequently, to obtain a prediction model, the data collected over a month and augmented to simulate the behavior during summer were used as the input to train a Graph Neural Network. After the evaluation of the results it was concluded that despite the considerable differences between the forecasted values and the reality, it was possible to perform such an implementation with limited data and resources. Moreover, a handful of traffic reporters still considered the results appropriate, and suitable.
Download
Paper Nr: 35
Title:

Modeling & Simulating the Evacuation of a Building Based on Building Floor Plan and Evacuation Strategies
Authors:

Shreya and K. S. Rajan
Abstract: Accidental fires in public and large buildings not only cause property loss but also can lead to loss of lives. During such emergencies, building evacuation depends on a range of factors including floor plans, exits available, obstructions if any, the occupancy levels of the building, and so on. The study here brings together the spatial, temporal, and path planning possibilities to evaluate fire evacuation strategies for 2D building plans. It provides a geospatial framework to assess the impacts of dynamic changes in the building environment and its impact on evacuation outcomes. In this study, occupancy-based path planning using Pgrouting over an IndoorGML formatted data is combined with modeling their interactions over the path toward the exit to assess the outcomes. This computational approach over the time-dependent path provides interesting insights into determining the number of paths and the need for one or more exits during an emergency. The study shows that integrating the floor plan into path generation and people flow can be a powerful tool for assessing the building environment.
Download