GISTAM 2020 Abstracts


Area 1 - Data Acquisition and Processing

Full Papers
Paper Nr: 13
Title:

Prediction of Spatiotemporal Distributions of Transient Urban Populations with Statistics Gathered by Cell Phones

Authors:

Toshihiro Osaragi and Ryo Hayasaka

Abstract: There is a growing demand for data that facilitate highly accurate understanding of the spatiotemporal distribution of both moving and static occupants in urban areas. Currently, a large amount of population data are available, however none of the data provide an accurate understanding of the numbers and departure/arrival locations of moving people using detailed units of space and time. In this paper, after evaluating the advantages and disadvantages of existing population statistics, including Mobile Spatial Statistics, Konzatsu-tokei®, and Person Trip survey data, we propose a method based on maximum likelihood method is investigated for using their strengths to best advantage and compensating for weaknesses. The proposed method is then validated by comparing with another flow data, which featured spatiotemporal data including departure/arrival locations, and demonstrate that the present procedure provides accurate estimates for population flows. This study makes it possible to analyse urban regions from new and never-before employed points of view by identifying the number of transient occupants and their travel directions at any time on high level of detail.
Download

Paper Nr: 14
Title:

Location Extraction from Twitter Messages using Bidirectional Long Short-Term Memory Model

Authors:

Zi Chen, Badal Pokharel, Bingnan Li and Samsung Lim

Abstract: Texts are a common form to encode location information which can be used crucially in disaster scenarios. While Named Entity Recognition (NER) has been applied to location extraction from formal texts, its performance on informal and colloquial texts such as social media messages is unsatisfactory. The geo-entities in social media are often neglected or categorized into unknown or ‘other’ entity types such as person or organisation. In this paper we proposed a Bidirectional Long Short-Term Memory (LSTM) Neural Netwok to identify location information especially aiming to recognize rarely known local places in social media messages. The contribution of both syntactic and semantic features to the classification results was explored as well. The proposed method was validated on a Twitter dataset collected from typhoon-affected areas, showing promising performance in detecting location information.
Download

Paper Nr: 61
Title:

Selfie Drones for 3D Modelling, Geological Mapping and Data Collection: Key Examples from Santorini Volcanic Complex, Greece

Authors:

Fabio L. Bonali, Varvara Antoniou, Othonas Vlasopoulos, Alessandro Tibaldi and Paraskevi Nomikou

Abstract: In the present work, we tested the use of selfie drones as a tool for 3D modeling, geological mapping, and data collection. The model we used is a 0.300-kg multirotor quadcopter being equipped with a 1/2.3-inch CMOS sensor capable of capturing 12 Megapixel pictures, attached to a 2-axis mechanical gimble and with approximately 16 minutes of flight time. Test sites are located in Santorini and are characterised by different settings: i) the 1570-1573 AD volcanic crater area, in Nea Kameni island, has a mostly horizontal topography; ii) the outcrop along Vlychada beach, showing layers of the Late Bronze Age (also well-known as Minoan) eruption, has mostly vertical topography. By applying the Structure from Motion techniques to pictures collected using the selfie drone, we were capable of: i) reconstructing the two sites with centimetric to sub-centimetric resolution; ii) recognizing geological features on very high-resolution Digital Surface Models and Ortomosaics; iii) mapping vertical cliffs made up of volcanic deposits on 3D Digital Outcrops Models; iv) collect new quantitative data for both sites.
Download

Short Papers
Paper Nr: 27
Title:

Use of Current Remote Sensing Methods for Biodiversity Monitoring and Conservation of Mount Kilimanjaro National Park Ecosystems

Authors:

Fortunata Msoffe, Thomas Nauss and Dirk Zeuss

Abstract: Climate and land use change have become serious challenges facing protected areas globally, more so those in the tropical forest ecosystems. Kilimanjaro-Mountain National Park was specifically designated to protect and safeguard the highest free-standing mountain in the tropics. The park attracts thousands of national and international tourists annually because of its snow capped-summit and the altitudinal gradients, representing the different eco-climatic zones of the world. Earnings from tourism boost the country’s economy while ensuring the sustainability of this unique glacial-tropical mountainous forest ecosystem park. Conventional monitoring of key biodiversity and environmental parameters are carried out by park staff, following established guidelines by Tanzania National Parks. However, given the park’s geo-morphological nature of mountainous terrain, efficient implementations of the labor intensive in-situ observations are hardly feasible. This research explored the use of Remote Sensing data from the European Satellite Agency– Sentinel-2 Multi-Spectral Instrument, in developing a state-of-the-art monitoring protocol. The developed methodology ensures that essential biodiversity parameters, including Vegetation Indices, required in monitoring the vast areas of the park and its surroundings in the short-term and long-term, using up to date, high resolutions and frequently available Remote Sensing data from the Sentinel-2 Sensors are captured.
Download

Paper Nr: 52
Title:

Quantifying Tufa Growth Rates (TGRs) using Structure-from-Motion (SfM) Photogrammetry

Authors:

Ivan Marić, Ante Šiljeg, Neven Cukrov and Fran Domazetović

Abstract: The production of high-quality digital surface models (DSMs) is an increasing interest throughout the various geomorphometry studies. Consequently, a wide range of advanced geospatial methods has been used at different scales. Despite the fact that Structure-from-Motion (SfM) photogrammetry is one of the most popular methods until now it has not been systematically applied in the studies of tufa formation dynamics (TFD). In this paper, we propose a framework for using SfM photogrammetry and GIS tools in the measurement of tufa growth rates (TGRs). TGRs were measured on two limestone plates (PLs) within the area of Roški waterfall in Croatia. Four submillimetre resolution DSMs of tufa have been created. TGR was 0.407 mm for a six-month period. Checkpoints were used to calculate errors. The results confirm the efficiency of the SfM at this scale. Research shows that photogrammetric measurement system design can produce extremely dense point clouds with high horizontal and vertical accuracy. The application of SfM and GIS in the measurement of TFD can be the great methodological improvement for specific geomorphometric applications at smaller scales.
Download

Paper Nr: 67
Title:

In-memory k Nearest Neighbor GPU-based Query Processing

Authors:

Polychronis Velentzas, Michael Vassilakopoulos and Antonio Corral

Abstract: The k Nearest Neighbor (k-NN) algorithm is widely used for classification in several application domains (medicine, economy, entertainment, etc.). Let a group of query points, for each of which we need to compute the k-NNs within a reference dataset to derive the dominating feature class. When the reference points volume is extremely big, it can be proved challenging to deliver low latency results. Furthermore, when the query points are originating from streams, the need for new methods arises to address the computational overhead. We propose and implement two in-memory GPU-based algorithms for the k-NN query, using the CUDA API and the Thrust library. The first one is based on a Brute Force approach and the second one is using heuristics to minimize the reference points near a query point. We also present an extensive experimental comparison against existing algorithms, using synthetic and real datasets. The results show that both of our algorithms outperform these algorithms, in terms of execution time as well as total volume of in-memory reference points that can be handled.
Download

Paper Nr: 71
Title:

Scale Drone Mapping on K8S: Auto-scale Drone Imagery Processing on Kubernetes-orchestrated On-premise Cloud-computing Platform

Authors:

Hemang Narendra Vithlani, Marcel Dogotari, Olee Hoi Ying Lam, Moritz Prüm, Bethany Melville, Frank Zimmer and Rolf Becker

Abstract: Aerial images acquired using drone-based imaging sensors can be processed by photogrammetry toolkits to create geometrically corrected 2D orthophoto and/or 3D models. This is a crucial step for many of the ever-evolving civil applications of drones such as precision agriculture and surveying. Nevertheless, limited computational resources become bottleneck in providing these results quickly. Cloud computing helps in such scenarios because of its value-added features, namely virtualization, elasticity, high performance and distributed computing for the web-based image processing. The containerization approach plays a vital role in cloud computing by providing operational efficiency. Container orchestration engine, Kubernetes, not only provides template-based or GUI-based service deployment but also better monitoring, log querying and auto-scaling. The present work displays a scalable photogrammetry service, deployed on a Kubernetes-orchestrated on-premise cluster. This reference implementation on Kubernetes enables the parallel processing of large datasets in less time than a single computer using the free and open-source toolkit OpenDroneMap.
Download

Paper Nr: 28
Title:

Towards Real Estate Analytics using Map Personalisation

Authors:

Mariam Mubarak, Kamran Khalid, Fizza Waqar, Ali Tahir, Ibraheem Haneef, Gavin McArdle and Michela Bertolotto

Abstract: The value of global real estate was $217 trillion in 2015 which is 2.7 times world GDP, making up roughly 60% of mainstream global assets and consequently it is considered one of the main drivers of economic growth. The availability of geospatial big data can assist real estate stakeholders to make informed decisions and increase their profits. Location plays a significant role in real estate decision making and so maps represent an excellent resource for real estate planning. Personalisation can assist with real estate decisions by ascertaining a user’s interests and preferences which can be captured via interaction with maps. A personalised real estate portal can then use this information to recommend properties on the web aiding property buyers and provide valuable real estate analytics. In this paper, we propose an approach for a personalised real estate platform called Estatech Maps. This will be a pioneer in the real estate industry, the key focus of which is to alter the prevailing management practices by imparting GIS and data analytics as long-term solutions.
Download

Paper Nr: 62
Title:

Urban Consumption Patterns: OpenStreetMap Quality for Social Science Research

Authors:

Hamidreza Rabiei-Dastjerdi, Gavin McArdle and Andrea Ballatore

Abstract: Citizen consumption refers to the goods and services which citizens utilise. This includes time spent on leisure and cultural activities as well as the consumption of necessary and luxury goods and services. The spatial dimension of consumption inequality can show the underlying urban spatial structure and processes of a city. Usually, the main barrier to effectively measuring consumption is the availability and accessibility of spatial data. While the main body of the literature utilises official, government data, such data is not always available, up-to-date or can be costly to acquire. In this paper, we discuss the potential of Volunteered Geographic Information (VGI) as a source of spatial data for determining consumption inequality. To this end, we compared OpenStreetMap (OSM) data, that can be used as proxies for consumption inequality, with official data in the area of Greater London. The results show that OSM is currently inadequate for studying the spatial dimension of consumption. It is our view that while VGI is appropriate for tasks such as routing and navigation, it also has the potential to add value to social science studies in the future.
Download

Area 2 - Remote Sensing

Full Papers
Paper Nr: 50
Title:

Application of Rodrigues Matrix in High Accuracy Geo-location for ZY-3 Panchromatic Imagery

Authors:

Xiaoming Gao, Fan Mo, Junfeng Xie and Qijun Li

Abstract: In this paper, Rodrigues matrix is proposed to establish constant angular error calibration model, and interior orientation errors are compensated by additional parameters model. Bundle block adjustment model is established by these two models on the basis of the rigorous geometric model for ZY-3 panchromatic imagery. Once the constant angular errors and interior orientation errors are eliminated using a few GCPs, the geo-location accuracy will be significantly improved.
Download

Short Papers
Paper Nr: 10
Title:

A Comparative Analysis of “Urban Expansion” using Remotely Sensed Data of CORINE Land Cover and Global Human Settlement Layer in Estonia

Authors:

Najmeh Mozaffaree Pour and Tõnu Oja

Abstract: Monitoring urban expansion is important because the policy makers in cities must detect the changes to provide services and manage resources for urban dwellers. In this study we analyse the built-up areas extracted from very high-resolution images of two important databases of CORINE land cover and GHSL; Built-Up Grid to map urban expansion at local level of cities of Tallinn and Tartu in their context (County) in Estonia. The reason for selecting these datasets was the representation of an available temporal data in many timespans which allowed extracting urban expansion in our case studies. The analysis was carried out over a subset of these datasets in ArcGIS environment and the data of GHSL-Built-Up Grid was extracted from Google Earth Engine platform. Therefore, the results showed that there was an increase in the amounts of built-up areas and its rate in these two counties while based on these two databases the results were not similar in areas and cells but similar in rate and growth patterns.
Download

Area 3 - Modeling, Representation and Visualization

Full Papers
Paper Nr: 59
Title:

Alas Landscape Modeling by Remote Sensing Image Analysis and Geographic Ontology: Study Case of Central Yakutia (Russia)

Authors:

Sébastien Gadal, Moisei Zakharov, Jūratė Kamičaitytė and Yuri Danilov

Abstract: Approaches of geographic ontologies can help to overcome the problems of ambiguity and uncertainty of remote sensing data analysis for modeling the landscapes as a multidimensional geographic object of research. Image analysis based on the geographic ontologies allows to recognize the elementary characteristics of the alas landscapes and their complexity. The methodology developed includes three levels of geographic object recognition: (1) the landscape land cover classification using Support Vector Machine (SVM) and Spectral Angle Mapper (SAM) classifiers; (2) the object-based image analysis (OBIA) used for the identification of alas landscape objects according to their morphologic structures using the Decision Tree Learning algorithm; (3) alas landscape’s identification and categorization integrating vegetation objects, territorial organizations, and human cognitive knowledge reflected on the geo-linguistic object-oriented database made in Central Yakutia. The result gives an ontology-based alas landscape model as a system of geographic objects (forests, grasslands, arable lands, termokarst lakes, rural areas, farms, repartition of built-up areas, etc.) developed under conditions of permafrost and with a high sensitivity to the climate change and its local variabilities. The proposed approach provides a multidimensional reliable recognition of alas landscape objects by remote sensing images analysis integrating human semantic knowledge model of Central Yakutia in the subarctic Siberia. This model requires to conduct a multitemporal dynamic analysis for the sustainability assessment and land management.
Download

Paper Nr: 74
Title:

A Pilot Project Proposal for the Implementation of a Geographic Information System for Immuno-Oncology in Italy

Authors:

Rosa M. Donolo, Paolo Collarile, Ilaria De Maria, Marta Donolo, Enrico Filippi, Maria Rizzo and Domenico Spagnolo

Abstract: The purpose of this project is to exploit the potential given by the connection of two fields apparently distant in which research is recently making very rapid progress: Immuno-Oncology (I-O) and Spatial Data Science. The connection of these two fields has led the research group to propose the building of an I-O Geographic Information System. In section 1 of this paper, we explain the advantages of linking I-O and Spatial Data Science, in section 2, we explain the purposes and the objectives of the project, in section 3, we describe the phases of the Geographical Information System (GIS) implementation and in section 4, we indicate some main issues and future perspectives. In particular, in section 3.1 we describe some preliminary steps in the building of the project’s database such as the collection of the Italian I-O Network Projects and the I-O excellence centres.
Download

Short Papers
Paper Nr: 40
Title:

From Pixels to 3D Representations of Buildings: A 3D Geo-visualization of Perspective Urban Respecting Some Urbanization Constraints

Authors:

Rani EL MEOUCHE, Mojtaba ESLAHI, Anne Ruas and Muhammad Ali A. SAMMUNEH

Abstract: In this paper, we generate the fictive 3D buildings and provide a 3D representation of an urban growth model using ArcGIS. SLEUTH urban growth model, like the other CA (Cellular Automata) models, creates a prospective 2D map containing some pixels on which urbanization is supposed to occur. These pixels have to be transformed into 3D building representations, while respecting some restrictions on urbanization. To create a building from a pixel, we transform the pixels from raster data to building footprints. In the process of transformation, different considerations and constraints are considered such as the direction of the footprints and the distances to urban objects and geographic features. To generate the 3D representations of the buildings, the appropriate heights are added to these footprints. The height of the buildings depends on the probability of the height of adjacent buildings. Although the provided 3D model is a primary and simple model, the 3D representation of the urban growth allows having different images of the city of tomorrow for supporting the scientists and authorities in charge of urban planner and management.
Download

Paper Nr: 53
Title:

GIScience Integrated with Computer Vision for the Interpretation and Analysis of Old Paintings

Authors:

Motti Zohar, Ilan Shimshoni and Fadi Khateb

Abstract: Photographs of Ottoman Palestine are available only from the 2nd half of the 19th century onward. Thus, in order to reconstruct the landscape at the time one should rely on other visual sources such as old paintings. To do so, their accuracy and completeness must be addressed first. In this paper we analyse a painting from 1823 by the British Sir Frederick Henniker that drew the Old City of Jerusalem when standing somewhere on the Mount of Olives. We use GIScience techniques with computer vision capabilities to resolve the exact location where the artist stood as well as verifying errors and completeness of the painting. Preliminary results demonstrate that the location of the artist when drawing the painting was on top of Mount of Olives (close to present-day 7 arches hotel) rather at the Cave of the Apostles as cited in the NLI librarian citation. Additionally, the accuracy of his work was verified by comparing the features he drew on the canvas to their actual location on a present-day photograph.
Download

Paper Nr: 55
Title:

Assessing the Vertical Accuracy of Worldview-3 Stereo-extracted Digital Surface Model over Olive Groves

Authors:

Fran Domazetović, Ante Šiljeg, Ivan Marić and Mladen Jurišić

Abstract: Worldview-3 stereo-extracted DSMs represent state-of-the-art products in the domain of satellite-based digital surface modelling. Main goal of our research was to evaluate the vertical accuracy of WV-3 derived DSMs over olive groves. Creation of high-accuracy WV-3 derived DSMs would allow efficient large scale management and protection of this valuable agricultural resource. Vertical accuracy of WV-3 derived DSM was evaluated at two test sites within Olive Gardens of Lun (Pag Island, Croatia), through the comparison with reference UAV photogrammetry derived VHR DSM. Two test sites were selected by object-based approach, established on spectral (NDVI, VARI) and height information (digital olive models (DOMs)). While first test site covers one single, individual oldest olive tree (45 m2), second test site covers larger area (2500 m2) with dense, unattended olive trees. Although vertical accuracy of individual olive trees still significantly deviates from reference model (RMSE = 3.604 m; MAE = 3.203 m), accuracy within larger test was much better (RMSE = 1.462 m; MAE = 1.127 m). This demonstrated that WV-3 stereo imagery has great potential for application in creation of DSMs over large scale forested areas, that would be hard to cover with field geospatial techniques (e.g. LiDAR or UAV photogrammetry).
Download

Paper Nr: 56
Title:

The Construction of a Network for Indoor Navigation

Authors:

Eliseo Clementini and Andrea Pagliaro

Abstract: Navigation systems provide help to moving agents by enabling them to reach a desired destination. The development of the Global Positioning System (GPS) has enabled the development of outdoor navigation systems, which are based on the knowledge of a map and the user’s location and provide guidance indications. The reality of indoor navigation systems is much different: difficulties stem from the fact that the movement of users is not constrained to belong to a well-established network such as the road network in the case of cars, but it is generally freer and less codifiable. This paper focuses on the automatic construction of a navigation graph superimposed on the map of a building that describes the possibilities of movement of a user within the building itself. The construction of the graph tries to replicate the spontaneous movement of users.
Download

Paper Nr: 57
Title:

Corinth Gulf Story Map: Enhancing Public Awareness in Natural and Anthropogenic Environment using Interactive GIS Applications

Authors:

Varvara Antoniou, Paraskevi Nomikou, Konstantinos Papaspyropoulos, Effrosyni Zafeirakopoulou, Othonas Vlasopoulos, Evangelia-Varvara Xrysopoulou, Eustathia Tziannou and Lemonia Ragia

Abstract: Story maps are widespread as an interactive tool used for science and spatial data communication, information and dissemination. A web-based application using story mapping technology is presented here to highlight places of interest around Corinth Gulf (Greece), a new addition in Natura 2000 areas. A tailored story map that combines thematic webmaps and scenes (3D webmaps) generated through a Geographic Information System (GIS) having a great impact on web-based visual presentations with narrative text and multimedia content was created to highlight the geological and cultural environment of the area around Corinth Gulf.
Download

Paper Nr: 65
Title:

Improving the Usability of the Land Cover and Use Information System of Spain (SIOSE): A Proposal to Distribute New Thematic Layers and Predefined Reclassifications

Authors:

Benito Zaragozí, José T. Navarro-Carrión, Jesús J. Rodríguez-Sala, Sergio Trilles and Alfredo Ramón-Morte

Abstract: Information on land use and land cover (LULC) is fundamental in the study and planning of human activities. In recent years, accessibility to quality geographical information has significantly increased, and this is also true for the case of LULC datasets. In Spain, the Land Cover and Use Information System of Spain (SIOSE) is concerned with harmonising access to this type of information through an object-oriented model and a series of technical specifications that regional administrations must follow. However, the information from SIOSE is so rich and complex that there is a usability gap that makes this data not exploited to its full potential in some contexts. In this communication, we analyse the context in which this usability gap occurs, its causes and consequences. Among other possible improvements, we suggest that enriching the SIOSE database with new thematic information would make its use more attractive and reduce the usability gap for less expert users. We propose an extension to the SIOSE object-oriented data model that will make it possible to enrich the LULC data with new data that are useful for various types of studies.
Download

Paper Nr: 66
Title:

Towards an Affordable GIS for Analysing Public Transport Mobility Data: A Preliminary File Naming Convention for Avoiding Duplication of Efforts

Authors:

Benito Zaragozí, Aaron Gutiérrez and Sergio Trilles

Abstract: Automated fare collection systems for public transport generate a large volume of information on the mobility of people in urban environments. New technologies associated with Big Data can facilitate the analysis of these data. However, the application of these technologies can be expensive and resource-demanding, especially in medium and small cities. This paper presents the case of the metropolitan transport authority of Tarragona, for which an affordable and extensible analysis system has been developed, based on relational databases and custom scripts. Among the technical problems that have had to be overcome, one of the first has been the unambiguous definition of the numerous queries required by mobility experts. For different reasons, mobility researchers request aggregate data queries from smart transport cards logs (e.g. providing a descriptive statement) and expect manageable tables to be analysed in a spreadsheet. To standardise the definition of queries, a domain-specific language as a file naming convention has been proposed with which database managers and mobility experts can communicate efficiently, avoiding confusion, duplication of efforts and other problems detected. The file naming convention has been applied as an early version within the defined use case to verify the viability of this idea.
Download

Paper Nr: 73
Title:

Optimal Path Planning for Drone Inspections of Linear Infrastructures

Authors:

Golizheh Mehrooz and Peter Schneider-Kamp

Abstract: Autonomous Beyond Visual Line of Sight (BVLOS) flights represent a huge opportunity in the drone industry due to their ability to monitor larger areas. Autonomous navigation and path planning are essential capabilities for BVLOS flights. In this paper, we introduce the routing component of a path planning system for inspecting linear infrastructures. We explore both a direct algorithm and a transformation algorithm. The direct algorithm is an extension of A* to allow limited routing through air as well as the use of non-logic intersections. The transformation algorithm pre-computes a graph that include edges for routing through air and nodes for non-logic intersections. We implemented both algorithms for routing along a particular type of linear infrastructure, power lines, and validated them through an empirical evaluation at three different scales: the Danish power grid, the French power grid, and the entire European power grid. The test results show that the transformation algorithm allows for sub-second routing performance for a small-to-medium sized power grid. Larger power grids can be routed in less than five seconds, and even an optimal route of more than six thousand kilometers along linear infrastructures from Portugal to Sweden via Russia is found in less than half a minute. All algorithms have been implemented and are available as an open-source Python package for Linear-infrastructure Mission Control (LiMiC).
Download

Paper Nr: 34
Title:

Analysis of Cartographic Generalization based on PYTHON Programming Language on Digital Topographic Maps

Authors:

Marko Stojanović, Siniša Drobnjak, Jasmina M. Jovanović, Nenad Galjak and Ana Vučićević

Abstract: Cartographic generalization is a creative process of abstraction, which is used in the design and content preparation of topographic maps. It includes the study of the geographic environment, processing of geographic data, and an evaluation with regard to type, purpose, and scale of the map, or selecting and merging their graphical presentation, with a big or small degree of abstraction. In the era of digital cartography more attention is paid to developing tools for automatic generalization of cartographic content. In this paper, automatic cartographic generalization is analyzed based on PYTHON programming language for production of digital topographic map scale 1:50 000 (DTM50) from digital topographic map scale 1:25 000 (DTM25).
Download

Paper Nr: 64
Title:

An Integrated Environmental Monitoring Approach through the Development of Coal Mine, a GIS Open Source Application

Authors:

L. Duarte, A. C. Teodoro, J. Fernandes, P. Santos and D. Flores

Abstract: Coal related fires may occur in un-mined outcrops, during coal mining, in abandoned mines, during storage and transportation and in coal waste deposits. The self-burning of coal mobilizes large amounts of pollutants, for instance, particulate matter, organic compounds and toxic trace elements that can be emitted, released or leached to soils, waters and air of the surrounding environment. The S. Pedro da Cova (Porto, Portugal) coal mine was exploited between 1795 and 1972 and had an important role on the economic development of the region. Nowadays a waste pile of about 28,000 m2 is still deposited in the mine, suffering from self-combustion since 2005. Geographical Information System (GIS) and spatial databases are frequently used for monitoring this type of processes. The main objective of this work was to integrate, manipulate and combine the spatial information obtained in the field with other datasets (geospatial and alphanumerical) in a GIS open source application connected to a relational database (PostGIS), in order to monitor and assess environmental conditions in the S. Pedro da Cova coal mine. This is an ongoing project where some campaigns were conducted and some spatial information was obtained (thermal images, Digital Elevation Model) and also water and soil samples.
Download

Area 4 - Knowledge Extraction and Management

Full Papers
Paper Nr: 15
Title:

A Cloud Architecture for Processing and Visualization of Geo-located 3D Digital Cultural Heritage Models

Authors:

Ikrom Nishanbaev

Abstract: The increasing affordability of surveying methods such as laser scanning and photogrammetry has aroused broad and current interest in 3D modelling among cultural heritage preservation specialists. This generated, in recent years, many digital cultural heritage preservation projects across the globe that aimed at documenting cultural heritage sites and objects in a 3D form. Once 3D cultural heritage models have been created, the next step is generally to assure their long-term digital storage, dissemination, and visualization. To this end, this article presents a new cloud architecture for processing and visualization of geo-located 3D cultural heritage models over the web, which has been accomplished by integrating maps, 3D cultural heritage models, and the geospatial data associated with the location of 3D cultural heritage models. The cloud architecture is based on Amazon Web Services, while the core framework for handling the content is managed by free and open-source, database-driven, easy-to-implement KeystoneJS Content Management System. All other frameworks used in the architecture such as for web mapping, 3D visualization, etc. are also based on free and open-source paradigm, which allows flexibility on extensions and re-use. The proposed architecture has been validated through a use-case applied to Australian 3D cultural heritage models.
Download

Paper Nr: 25
Title:

Combining Spatial Data Layers using Fuzzy Inference Systems: Application to an Agronomic Case Study

Authors:

Serge Guillaume, Terry Bates, Jean-Luc Lablée, Thom Betts and James Taylor

Abstract: This paper presents an application of Fuzzy Logic, well known for its linguistic modeling ability, in a multi-criteria decision making framework applied to spatial data sets. The Fuzzy Logic is integrated in two different ways. First, fuzzy sets are used to model an expert preference relation for each of the individual spatial information sources to turn raw data into satisfaction degrees. Second, fuzzy rules are used to model the interaction between sources to aggregate the individual degrees into a global score. The whole framework is implemented in an open source software called GeoFIS. The potential of the method is illustrated using a typical farming decision: the design of a nitrogen fertilization map within a vineyard. The vineyard is a Concord (Vitis labrusca) juice grape vineyard in the Lake Erie region of New York state. The vineyard manager and a local research/extension viticulturist both used the tool to generate a prescription nitrogen map based on their knowledge and spatial crop and soil information. The process captured different preferences between the two users (industry vs. research) and generated different prescription maps that reflected their differing objectives, knowledge and risk perception in vine management. Although applied to vineyard data, this decision tool has a wide potential application to agri-environmental (and other) spatial data sets.
Download

Paper Nr: 43
Title:

Spatial Entity Resolution between Restaurant Locations and Transportation Destinations in Southeast Asia

Authors:

Emily Gao and Dominic Widdows

Abstract: As a tech company, Grab has expanded from transportation to food delivery, aiming to serve Southeast Asia with hyperlocalized applications. Information about places as transportation destinations can help to improve our knowledge about places as restaurants, so long as the spatial entity resolution problem between these datasets can be solved. In this project, we attempted to recognize identical place entities from databases of Points-of-Interest (POI) and GrabFood restaurants, using their spatial and textual attributes, i.e., latitude, longitude, place name, and street address. Distance metrics were calculated for these attributes and fed to tree-based classifiers. POI-restaurant matching was conducted separately for Singapore, Philippines, Indonesia, and Malaysia. Experimental estimates demonstrate that a matching POI can be found for over 35% of restaurants in these countries. As part of these estimates, test datasets were manually created, and RandomForest, AdaBoost, Gradient Boosting, and XGBoost perform well, with most accuracy, precision, and recall scores close to or higher than 90% for matched vs. unmatched classification. To the authors’ knowledge, there are no previous published scientific papers devoted to matching of spatial entities for the Southeast Asia region.
Download

Paper Nr: 72
Title:

A GIS Open Source Application to Perform the Spatial Distribution of Prevention Quality Indicators (PQIs)

Authors:

L. Duarte, M. Lobo, J. Viana, A. Freitas and A. C. Teodoro

Abstract: Geographical variations carry important information for improving and planning more equitable and sustainable health care services. Geographic Information Systems (GIS) are crucial tools that provide intuitive visual help which contributes to a better understanding of the spatial distribution of health risk factors, resources, care and outcomes. The interest in GISs have stimulated the development of several applications worldwide to publicly inform the geographical patterns of health. However, in Portugal, this type of tools remains underdeveloped for public reporting of health information. The aim of this study was to develop a GIS open source application for spatial analysis of healthcare indicators in Portugal, using hospital data obtained from the Administração Central do Sistema de Saúde, I.P. Specifically, given their importance to monitor the quality of primary health care, data regarding Prevention Quality Indicators (PQIs) will be used to establish a proof of concept of this tool. The tool was connected to a spatial database in order to filter the parameters. Several maps based on PQI information were created in order to test the application. It was concluded that the spatial combination of all the data provided in a GIS software and through an intuitive application can contribute to the analysis of quality of primary health care.
Download

Short Papers
Paper Nr: 17
Title:

Geolocation Prediction from Tweets: A Case Study of Influenza-like Illness in Australia

Authors:

Bingnan Li, Zi Chen and Samsung Lim

Abstract: Twitter has become an effective platform for gathering massive event-related data from growing popularity. It provides an approach to monitoring and analysis of the emergence and devolvement of events. In the field of data mining and social media analysis, geographic information is an important element to be factored in. However, only nearly 2% of tweets contain accurate geographic information because of various concerns e.g. complexity and privacy. In order to overcome this restriction, devising methods of geolocation prediction has become the main topic in this filed. Geographic information plays a valuable role in responding to the control and surveillance of epidemic diseases. In this study, we constructed a geolocation prediction method based on potential location-related tweet metadata. Coordinate information can be calculated from the bounding box, while location information can be extracted from the text content, the user’s location at the time of use and the labelled place names using the Named Entity Recognition technique. Three types of coordinate sets of Australian suburbs are defined and used to construct coordinates references from the place names. Models with different parameters have been applied to predict geolocations of influenza-like illness from the tweets of the 2019 flu season in Australia. The results show that the proposed models with four parameters perform better than the existing models. When the area threshold is set to 4,500 km2, the best model can successfully predict influenza-like illness with the mean error distance of 4.65 km and the median error distance of 2.57 km. Hence the proposed method is shown to enhance the geographic information associated with the tweets and make the emergency response to influenza-like illness more effective and efficient.
Download

Paper Nr: 21
Title:

Occupancy Grid Map Generation from OSM Indoor Data for Indoor Positioning Applications

Authors:

Thomas Graichen, Rebecca Schmidt, Julia Richter and Ulrich Heinkel

Abstract: In recent years, there is a growing interest in indoor positioning due to the increasing amount of applications that employ position data. Current approaches determining the location of objects in indoor environments are facing problems with the accuracy of the sensor data used for positioning. A solution to compensate inaccurate and unreliable sensor data is to include further information about the objects to be positioned and about the environment into the positioning algorithm. For this purpose, occupancy grid maps (OGMs) can be used to correct such noisy data by modelling the occupancy probability of objects being at a certain location in a specific environment. In that way, improbable sensor measurements can be corrected. Previous approaches, however, have focussed only on OGM generation for outdoor environments or require manual steps. There remains need for research examining the automatic generation of OGMs from detailed indoor map data. Therefore, our study proposes an algorithm for automated OGM generation using crowd-sourced OpenStreetMap indoor data. Our experiments with nine different building map datasets demonstrate that the proposed method provides reliable OGM outputs. The proposed algorithm now enables the integration of environmental information into positioning algorithms to finally increase the accuracy of indoor positioning applications.
Download

Paper Nr: 45
Title:

In-house Localization for Wi-Fi Coverage Diagnostics

Authors:

Filipe Meneses, Ricardo Ferreira, Adriano Moreira and Carlos M. Martins

Abstract: Telecommunication operators and Internet Service Providers often face the problem of having residential customers complaining about deficient Wi-Fi coverage inside their houses and/or about the low quality of service while accessing the Internet. Addressing these complaints properly involves a comprehensive in-house diagnostic of the technical deployment, the use of specialized equipment and visits by qualified personnel. An alternative is to involve the users in a preliminary diagnostic, by leveraging the potential of current smartphones, aiming to identify possible causes for the complaints that can be solved remotely or through simple procedures to be executed by the customers. A key feature of such a diagnostic procedure is the ability to estimate the location of the smartphone indoors automatically. This paper proposes a simple indoor localization solution, based on Wi-Fi fingerprinting, that can be integrated into one such diagnostics procedure. The proposed solution was implemented and tested in real-world houses by emulating the behaviour of non-qualified users. The obtained results show that Wi-Fi fingerprinting, when used in such an uncontrolled environment, still poses some challenges as its precision is still significantly low.
Download

Paper Nr: 54
Title:

Multiscale GIS based Analysis of Urban Green Spaces (UGS) Accessibility: Case Study of Sisak (Croatia)

Authors:

Silvija Šiljeg, Rina Milošević and Edita Vilić

Abstract: By the year 2050, two-thirds of the world population will live in urban areas. Therefore the quality of life in cities has become the object of numerous research papers. One of the basic elements of satisfying the quality of life is the accessibility of urban green spaces (UGS). In this paper accessibility of UGS for the city of Sisak (Croatia) has been analysed. Based on the fact that Sisak is traditionally an industrial type of town, the optimal distribution of UGS has the potential to ease the negative effects of urbanization and industrialization. Accessibility analysis was performed according to guidelines of ANGst (Accessible Natural Greenspace Standard) methodology. The research is conducted at a multiscale level (based on GIS analysis). The primary spatial database of UGS for the city of Sisak was created using the supervised classification method of Sentinel-2A images and vectorization of high-resolution digital orthophoto (DOP). Accessibility zones were generated using the Network Analyst extension. Results show that the basic ANGst standard of UGS accessibility is not satisfactory throughout the city. To get more detailed results we suggest using the very high-resolution satellite imagery or aerial photogrammetry.
Download

Paper Nr: 58
Title:

Assessment of the Completeness of OpenStreetMap and Google Maps for the Province of Pavia (Italy)

Authors:

Marica Franzini, Laura Annovazzi-Lodi and Vittorio Casella

Abstract: Free access web-based mapping is nowadays largely used in several areas such as navigation, location-based services or when it is necessary to obtain quickly geographical information. Some of them are based on volunteers’ work, among which OpenStreetMap (OSM), while some others were design for commercial purposes, such as Google Maps (GM). Given the variety of contributors and their heterogeneity, one of the critical aspects of OSM is the homogeneity and quality level of its information; furthermore, GM is also largely consulted but presents inhomogeneity between densely populated and rural areas. The paper aims at analysing the buildings completeness of OSM and GM over the Province of Pavia, in Northern Italy: the applied method will be presented together with the results obtained at two different time frames (spring 2018 and winter 2018). Finally, a quick review about the volunteers that had effectively contributed to OSM will be presented.
Download

Area 5 - Domain Applications

Full Papers
Paper Nr: 4
Title:

A Public Participatory Approach toward the Development of a Comprehensive Geospatial Database in Support of High-scale Food Security Analysis

Authors:

Timothy Mulrooney and Tysean Wooten

Abstract: While Geographic Information Systems (GIS) has slowly been integrated into the study of the food environment, little research has been performed to determine the data development needs and standards that best necessitate high-quality research at a high scale. In an era with limited resources such as personnel, bandwidth, space and time, the optimization of these resources in order to understand, visualize and facilitate interventions at an appropriate scale is critical if not necessary. In this research, subject matter experts assessed and evaluated the relative importance of various GIS data themes, attributes and facets of GIS database development in support of local-scale food security analysis. It was found that factors related to the placement of various food sources (grocery stores and farmers markets) and individualized vehicular transportation (roads) outweighed those related to land cover, utilities and zoning, as well as non-vehicular (sidewalks) and public (bus routes) means of transportation. In addition, when ranking various dimensions of data quality, subject matter experts found positional accuracy and attribute accuracy to be the most important when undertaking the development of a geospatial database of this magnitude.
Download

Paper Nr: 32
Title:

Design and Development of an Application for Predicting Bus Travel Times using a Segmentation Approach

Authors:

Ankhit Pandurangi, Clare Byrne, Candis Anderson, Enxi Cui and Gavin McArdle

Abstract: Public transportation applications today face a unique challenge: Providing easy-to-use and intuitive design, while at the same time giving the end user the most updated and accurate information possible. Applications often sacrifice one for the other, finding it hard to balance the two. Furthermore, accurately predicting travel times for public transport is a non-trivial task. Taking factors such as traffic, weather, or delays into account is a complex challenge. This paper describes a data driven analysis approach to resolve this problem by using machine learning to estimate the travel time of buses and places the results in a user-friendly application. In particular, this paper discusses a predictive model which estimates the travel time between pairs of bus stops. The approach is validated using data from the bus network in Dublin, Ireland. While the evaluation of the predictive models show that journey segment predictions are less accurate than the prediction of a bus route in full, the segmented approach gives the user more flexibility in planning a journey.
Download

Paper Nr: 35
Title:

Relation between Proximity to Public Open Spaces and Socio-economic Level in Three Cities in the Ecuadorian Andes

Authors:

María L. Guerrero, Daniel Orellana, Jorge Andrade and Gabriela Naranjo

Abstract: Public Open Spaces (POSs) are necessary urban goods for satisfying personal and collective needs for physical, social and mental wellbeing. Equitable spatial access to POSs is key for guaranteeing that resources for wellbeing are democratically available for all members of the community. Environmental justice states that contemporary cities have a biased distribution of public spaces, against socially and economically more disadvantaged sectors of society. Under these premises, this paper evaluates whether there is a case of environmental imbalance in access to public spaces in three Ecuadorian cities: Quito, Cuenca and Ibarra, based on the socio-economic status of the population. A pedestrian impedance street network model was used for obtaining time to the nearest Public Open Space from each urban block, and socio-economic conditions were obtained from national census data per household and divided into quartiles. Statistical analyses included Mood’s Median Test, Dunn’s post-hoc test and notched boxplots for assessment. Results show that there is a significant difference in time to public spaces between quartiles, where the quartile with the lowest socioeconomic conditions is also further from public spaces than the others in the three cities. These results should inform planning policies, strategies, designs and decisions for future leisure land use reserves.
Download

Short Papers
Paper Nr: 12
Title:

Forecasting Travel Times with Space Partitioning Methods

Authors:

Jhonny Pincay, Alvin O. Mensah, Edy Portmann and Luis Terán

Abstract: Roads and streets are more and more crowded. For delivery companies that use road transportation, this is a concerning issue as longer times spent on roads mean higher operational costs and less customer satisfaction. Nevertheless, the data captured during operation hours of their vehicles can be leveraged to address such issues. This, however, is not a straightforward task given the possible low number of vehicles covering one route and the complexities introduced by the delivery business nature. The present research work proposes an approach to forecast travel time through the use of probe data from logistic vehicles and simple mathematical models. The delivery operations of five months of a vehicle from the Swiss Post, the national postal service company of Switzerland, were studied in a segment-to-segment manner, following a four-step method. Moreover, the results of the forecasting were evaluated calculating the mean absolute percentage error and mean absolute error metrics. The results obtained indicate that is possible to achieve a considerable forecasting accuracy without the deployment of a large number of vehicles or the implementation of complex algorithms.
Download

Paper Nr: 44
Title:

Aiding Irrigation Census in Developing Countries by Detecting Minor Irrigation Structures from Satellite Imagery

Authors:

Chintan Tundia, Pooja Tank and Om Damani

Abstract: Minor irrigation structures such as well and farm ponds play very important roles in agriculture growth in developing countries. Typically, a minor irrigation census is conducted every five years to take inventory of these structures. It is essential that an up to date database of these structures be maintained for planning and policy formulation purposes. In this work, we present the design and implementation of an online system for the automatic detection of irrigation structures from satellite images. Our system is built using three popular object detection architectures - YOLO, FasterRCNN and RetinaNet. Our system takes input at multiple resolutions and fragments and reassembles the input region to perform object detection. Since currently there exists no dataset for farm pond and the only publicly available well dataset covers a small geographical region, we have prepared object detection datasets for farm ponds and wells using Google Maps satellite images. We compare the performance of a number of state of the art object detection models and find that a clear trade-off exists between the detection accuracy and inference time with the RetinaNet providing a golden mean.
Download