Airborne Light Detection and Ranging (LIDAR) technology has become the primary method to derive high-resolution Digital Terrain Models (DTMs), which are essential for studying Earth’s surface processes, such as flooding and landslides. The critical step in generating a DTM is to separate ground and non-ground measurements in a voluminous point LIDAR dataset, using a filter, because the DTM is created by interpolating ground points. As one of widely used filtering methods, the progressive morphological (PM) filter has the advantages of classifying the LIDAR data at the point level, a linear computational complexity, and preserving the geometric shapes of terrain features. The filter works well in an urban setting with a gentle slope and a mixture of vegetation and buildings. However, the PM filter often removes ground measurements incorrectly at the topographic high area, along with large sizes of non-ground objects, because it uses a constant threshold slope, resulting in “cut-off” errors. A novel cluster analysis method was developed in this study and incorporated into the PM filter to prevent the removal of the ground measurements at topographic highs. Furthermore, to obtain the optimal filtering results for an area with undulating terrain, a trend analysis method was developed to adaptively estimate the slope-related thresholds of the PM filter based on changes of topographic slopes and the characteristics of non-terrain objects. The comparison of the PM and generalized adaptive PM (GAPM) filters for selected study areas indicates that the GAPM filter preserves the most “cut-off” points removed incorrectly by the PM filter. The application of the GAPM filter to seven ISPRS benchmark datasets shows that the GAPM filter reduces the filtering error by 20% on average, compared with the method used by the popular commercial software TerraScan. The combination of the cluster method, adaptive trend analysis, and the PM filter allows users without much experience in processing LIDAR data to effectively and efficiently identify ground measurements for the complex terrains in a large LIDAR data set. The GAPM filter is highly automatic and requires little human input. Therefore, it can significantly reduce the effort of manually processing voluminous LIDAR measurements.

VIII Conferencia Internacional BIREDIAL – ISTEC 2018 22 al 25 de octubre de 2018. Organizado por la Pontificia Universidad Católica del Perú. Lima Perú / La Universidad Peruana de Ciencias Aplicadas (UPC) ha desarrollado, desde finales del año 2017, diversas iniciativas para generar una infraestructura de servicios geoespaciales para sus tesis y otros tipos de producción intelectual, con los siguientes fines: mejorar su accesibilidad; implementar mapas temáticos; identificar nuevas formas de difusión en el marco de los repositorios académicos; y promover el intercambio de los datos y servicios de información espacial a nivel local e internacional mediante estándares: Norma ISO 19115, Content Standard for Digital Geospatial Metadata (CSDGM) y el estándar Open Geospatial Consortium (OGC). Esta iniciativa guarda relación con diversas acciones para el fortalecimiento de una infraestructura de datos espaciales desplegadas por el Estado Peruano. En este esfuerzo, se promulga la Resolución Ministerial N° 126-2003-PCM, que constituye el Comité Coordinador de la Infraestructura de Datos Espaciales del Perú (CC-IDEP) y el Decreto Supremo 133-2013-PCM, el cual establece como obligatorio el acceso e intercambio de información espacial entre entidades de la administración pública y promueve la creación de infraestructuras de datos espaciales institucionales. En este contexto, las iniciativas de la UPC se han enfocado en: 1. Generación de mapas temáticos en Psicología y Arquitectura a partir del Repositorio Académico UPC. 2. Modificación del Reglamento de las tesis para agregar la información geoespacial en los metadatos del Repositorio Académico UPC (ubicación geográfica y/o UTM). 3. Desarrollo de un análisis cualitativo de las tesis en Psicología mediante mapas temáticos de tres repositorios institucionales peruanos. 4. Utilización de softwares de sistemas de información geográfica (SIG) para generar reportes de mapas temáticos. 5. Desarrollar estrategias de Search Engine Optimization (SEO) para mejorar la visibilidad de la información geoespacial. Las siguientes acciones se orientan: 1. Generación de mapas temáticos de tesis en todos los programas académicos a partir del Repositorio Académico. 2. Evaluación de la visibilidad del Repositorio Académico UPC a partir de la generación de mapas temáticos. 3. Establecer recomendaciones para la generación de servicios geoespaciales en las Bibliotecas. 4. Establecer convenios con organismos especializados en información geoespacial a nivel local e internacional.

Stora mängder miljödata samlas hela tiden in och för att använda all data behöver den förstås av användarna så de kan applicera kunskapen inom deras område. Visualisering skapar förståelse om data. Heatmaps kan användas för att visualisera geospatial data och interaktivitet är ett hjälpmedel för att skapa ytterligare grafiska representationer. I detta arbete evalueras JavaScript-teknikerna D3.js, Heatmap.js och Vue.js angående vad som är mest lämpligt för att visualisera geospatial data utifrån effektiviteten vid utritning av heatmaps. Ett experiment genomförs där biblioteken D3.js, Heatmap.js testas i ramverket Vue.js. Detta för att ta reda på vilket bibliotek som föredras vid utritning av heatmaps och om ett ramverk påverkar resultatet. En miljö sätts upp för att genomföra undersökningen och tester för att påvisa detta. Resultatet indikerar att Heatmap.js och mindre datamängder ger en lägre utritningstid i den tillämpning som undersökts. I framtiden är det intressant att undersöka flera bibliotek och flera datamängder.

Deep learning ship detection in satellite optical imagery suffers from false positive occurrences with clouds, landmasses, and man-made objects that interfere with correctly classifying ships. A custom U-Net is implemented to challenge this issue and aims to capture more features in order to provide a more accurate class accuracy. This model is trained with two different systematic architectures: single node architecture and a parameter server variant whose workers act as a boosting mechanism. To ex-tend this effort, a refining method of offline hard example mining aims to improve the accuracy of the trained models in both the validation and target datasets however it results in over correction and a decrease in accuracy. The single node architecture results in 92% class accuracy over the validation dataset and 68% over the target dataset. This exceeds class accuracy scores in related works which reached up to 88%. A parameter server variant results in class accuracy of 86% over the validation set and 73% over the target dataset. The custom U-Net is able to achieve acceptable and high class accuracy on a subset of training data keeping training time and cost low in cloud based solutions. / Graduate

Ett teknikorienterat experiment genomfördes innehållande jämförelsen av renderingstid för datavisualisering av geospatial demografisk data med SVG och Canvas. Med ökande datamängder ökar behovet för hantering, lagring och analysering av den. För att underlätta tolkning av den kan visualiseringstekniker tillämpas. En vanlig visualiseringsteknik för demografisk data är koropletkartor. där specifika områden färgläggs för att beskriva den demografiska spridningen. Ramverk granskades, D3.js användes för SVG-rendering och ECharts för Canvas-rendering. Problemet är att svarstider är direktkopplade till användarupplevelsen och riskerar att försämras vid visualisering av större datavolym i samband med SVG. Därav undersöktes om visualisering av olika mängder data i samband med Canvas kan vara en lämplig lösning på problemet. Studien sammanfattade att SVG lämpade bättre för enkla-, till medelkomplexa kartsorter medan ingen skillnad kunde ses mellan renderingsteknikerna vid rendering av komplexa koropletkartor. Vidare bör människoorienterade experiment, fler teknologier samt diverse sorters data studeras för att öka bredden, generaliserbarheten samt utesluta påverkande faktorer.

This paper aims to explore the perceptual recognition of topographical building patterns from real-world OpenStreetMap data on virtual globes. An implementation was developed in which all geographical and contextual information was layered and, for the purpose of this study, what solely remained were building patterns as viewed from above. This was developed as a module for the planetarium visualization software Uniview. The aim was to determine how cities with different building patterns were perceived by participants in terms of size, scale, and building density. This was measured as the comparative difference between city pairs, that is, how much they differed in the percentage of the area that they covered. Two quantitative studies were conducted, one smaller controlled study with 19 participants and one larger online crowd-sourced study with 72 participants. The results show that participants are generally able to discern building patterns when the comparative difference is greater than a certain critical threshold. This critical threshold was determined to be at approximately 0.5% for both studies and for accuracy levels above 60%. Thus it was concluded that below this critical threshold users should be provided with visual feedback or other means of identifiers in order to allow for definite recognition, depending on what kind of information a certain type of visualization is trying to convey. / Den här rapporten avser att utforska den perpetuella igenkänningen av topografiska byggnadsmönster genom att använda geografisk data från OpenStreetMap som avbildas på virtuella sfärer. En implementation utvecklades där geografisk data samt kontextuell information ordnades i överlappande lager som filtrerades, och där endast byggnadsmönster sett från ovan kvarstod. Denna modul utvecklades för Uniview som är en mjukvara för visualisering i planetarier. Målet var att avgöra hur deltagare uppfattade städer med olika byggnadsmönster med hänsyn till storlek, skala, samt byggnadsdensitet. Detta mättes genom den procentuella skillnaden mellan städer, dvs. skillnaden i procent för varje stads geografiska utsträckning. Två kvantitativa studier utfördes, en mindre kontrollerad studie med 19 deltagare samt en större nätbaserad studie med 72 deltagare. Resultatet visar att deltagare generellt kunde bedöma den procentuella skillnaden i byggnadsmönster upp till en viss kritisk gräns. Denna kritiska gräns fastställdes till runt 0.5% för båda studier och för noggrannhetsnivåer över 60%. Slutsatsen från detta är att användare bör ges visuella indikatorer för nivåer under denna kritiska gräns för att säkerställa definitiv igenkänning beroende på vilken information som skall förmedlas i en viss typ av visualisering.

The use of databases for managing spatial data is widespread due to the efficiency of traditional SQL databases like Azure SQL. However, the exponential growth of data from sources like social media has led to the popularity of NoSQL databases such as MongoDB that handle large volumes of data effectively. NoSQL databases, including MongoDB, have built-in support for geospatial queries, making them suitable for managing geospatial data. Geospatial data combines geometric and geographic information and is represented by spatial datatypes like Point, LineString, and Polygon. MySQL and MongoDB both support geospatial data, but limited studies are comparing their performance in geospatial queries. An experiment was conducted to compare the fetch speed of geospatial data in these databases. The results were analyzed using graphs and related studies to draw conclusions, which showed that MongoDB performed slower fetch requests than MySQL. Future studies can use more data points and different queries.

The advent of Intelligent Transportation Systems (ITS) and smart cities, powered by sensors and cyber-physical systems, has transformed urban mobility planning through data-driven approaches. Advances in communication technologies enable the collection of large-scale mobility data, offering valuable insights into mobility patterns within urban road networks. Key sources of this data, such as vehicular travel durations and driver trajectory behaviors, are crucial for understanding the dynamics of traffic flow and drivers' interactions with urban road systems. This dissertation presents geospatial probabilistic machine learning models designed to capture spatiotemporal and contextual properties of vehicular mobility patterns in urban environments. It further emphasizes how these insights can address operational challenges, particularly in emergency response systems, where ambulances interact closely with urban road networks. Optimizing decision-making in such systems is intricately linked to efficient navigation, accurate travel duration prediction, and effective routing, all of which are deeply tied to understanding mobility patterns and dynamics. By its stochastic nature, mobility data is inherently uncertain, dynamic, sparse, and noisy, influenced by diverse spatiotemporal and exogenous factors. Probabilistic models are particularly well-suited for addressing these challenges, as they effectively handle uncertainty, variability, and noise, and their extensions, in the right way, are capable of handling sparse information and dynamic conditions. This dissertation focuses on probabilistic models, emphasizing their robustness and ability to generalize to new scenarios and cities, making them a powerful tool for effectively learning urban mobility dynamics and enhancing transportation systems' resilience and sustainability.

Context: Accurate decision-making requires updated and precise information to establish the reality of an overall situation. New data sources (e.g., wearable technologies) have been increasing the amount of available and useful data, which is now called big data. This has a great potential for transforming the entire business process and improving the accuracy of decisions. In this context, disaster management represents an interesting scenario that relies on big data to enhance decision-making. This is because it must cope with data provided not only by traditional sources (e.g., stationary sensors) but also by emerging sources - for instance, information shared by local volunteers, i.e., volunteered geographic information (VGI). When combined, these data sources can be regarded as large in volume, with different velocities, and a variety of formats. Furthermore, an analysis is required to confirm their veracity is required since these data sources are disconnected and prone to various errors. These are the 4Vs that characterize big data. Gap: However, although all these data open up further opportunities, their huge volume, together with an inappropriate data integration and unsuitable visualization, can result in information being overlooked by decision-makers. This problem arises because the integration of the available data is hampered by the intrinsic heterogeneity of their features (e.g., their occurrence in different formats). When integrated, this information also often fails to reach the decision-makers in a suitable way (e.g., in appropriate visualization formats). Moreover, there is not a clear understanding of the decision-makers needs or how the available data can meet these needs. Objective: In light of this, this thesis presents an approach for improving decision-making with heterogeneous geospatial big data based on spatial decision support systems and volunteered geographic information in disaster management. Methods: Systematic mapping studies were conducted to identify gaps in research studies with regard to the use of volunteered information and spatial decision support systems in disaster management. On the basis of these studies, two design science projects were carried out. The first of these aimed at defining the elements that are essential for ensuring the integration of heterogeneous data, whereas the second project aimed at obtaining a better understanding of decision-makers needs. A cross-organizational action research project was also conducted to define the design principles that should be observed for a spatial decision support system to effectively support decision-making with heterogeneous geospatial big data. A series of empirical case studies was undertaken to evaluate the outcomes of these projects. Results: The overall approach thus consists of the three significant outcomes that were derived from these projects. The first outcome was the conceptual architecture that defines the integration of heterogeneous data sources. The second outcome was a model-based framework that describes the connection of decision-making with appropriate data sources. The third outcome is based on the framework and comprises a set of design principles for guiding the development of spatial decision support systems for decision-making with heterogeneous geospatial big data. Conclusion: This thesis has made a useful contribution to both practice and research. In short, it defines ways of integrating heterogeneous data sources, provides a better understanding of decision-makers needs, and supports the development of a spatial decision support system to effectively assist decision-making with heterogeneous geospatial big data. / Contexto: Uma tomada de decisão precisa exige informações mais precisas e atualizadas para estabelecer a realidade da situação geral. Novas fontes de dados (e.g, tecnologias vestíveis) tem aumentado a quantidade de dados úteis disponíveis, que agora é chamado de big data. Isso tem grande potencial para transformar todo o processo de negócio e melhorar a precisão na tomada de decisão. Neste contexto, a gestão de desastres representa um interessante cenário que depende de big data para aprimorar a tomada de decisão. Isso porque, ela tem que lidar com dados fornecidos não apenas por fontes tradicionais (e.g., sensores estáticos), mas também por fontes emergentes por exemplo, informações compartilhadas por voluntários locais, i.e., as informações geográficas de voluntários (VGI). Quando combinadas, estas fontes de dados podem ser consideradas grandes em volume, com diferentes velocidades e uma variedade de formatos. Além disso, uma análise com relação à sua veracidade é necessaria uma vez que estas fontes de dados são desconectadas e propensas à erros. Estes são os 4Vs que caracterizam big data. Problema: No entanto, embora todos estes dados abrem novas oportunidades, seu grande volume em conjunto com uma integração inapropriada e uma visualização inadequada, podem tornar as informações ignoradas por tomadores de decisão. Isso ocorre, pois, a integração dos dados disponíveis torna-se complicada devido a heterogeneidade intrínseca nas suas características (e.g., dados em formatos diferentes). Quando integradas, estas informações frequentemente também não chegam aos tomadores de decisão em uma condição apropriada (por exemplo, no formato de visualização adequado). Além disso, não existe uma clara compreensão sobre as necessidades dos tomadores de decisão ou sobre como os dados disponíveis podem ser usados para atender essas necessidades. Objetivo: Dessa forma, esta tese de doutorado apresenta uma abordagem para melhorar a tomada de decisões com grande volume de dados espaciais heterogêneos baseada em sistemas de suporte à decisão espacial e informações geográficas de voluntários na gestão de desastres. Métodos: Mapeamentos sistemáticos foram conduzidos para identificar lacunas de pesquisa no uso de dados voluntários e sistemas de suporte à decisão na gestão de desastres. Com base nestes estudos, dois projetos de design science foram conduzidos. O primeiro deles buscou definir elementos essências para entender a integração de dados heterogêneos, enquanto o segundo projeto buscou fornecer um melhor entendimento das necessidades dos tomadores de decisão. Também foi conduzido um projeto de pesquisa-ação interinstitucional para definir princípios de projeto que deveriam ser observados para um sistema de suporte à decisão espacial ser efetivo no apoio a tomada de decisão com grande volume de dados espaciais heterogêneos. Uma série de estudos de caso empíricos foram conduzidos para avaliar os resultados destes projetos. Resultados: A abordagem geral então é composta pelos três resultados significantes que foram derivados destes projetos. Em primeiro lugar, uma arquitetura conceitual que especifica a integração de fontes de dados heterogêneas. O segundo elemento é uma estrutura baseada em modelo que descreve a conexão entre a tomada de decisão com as fontes de dados mais adequadas. Com base nesta estrutura, o terceiro elemento consiste em um conjunto de princípios de design que guiam o desenvolvimento de um sistema de suporte à decisão espacial para tomada de decisão com grande volume de dados espaciais heterogêneos. Conclusão: Esta tese de doutorado realizou importantes contribuições para a prática e pesquisa. Em resumo, ela define formas para integrar fontes de dados heterogêneos, fornece uma melhor compreensão sobre as necessidades dos tomadores de decisão e ajuda no desenvolvimento de sistemas de suporte à decisão espacial para tomada de decisão com grande volume de dados espaciais heterogêneos.

Human activities in the past century have caused a variety of environmental problems in South Florida. In 2000, Congress authorized the Comprehensive Everglades Restoration Plan (CERP), a $10.5-billion mission to restore the South Florida ecosystem. Environmental projects in CERP require salinity monitoring in Florida Bay to provide measures of the effects of restoration on the Everglades ecosystem. However current salinity monitoring cannot cover large areas and is costly, time-consuming, and laborintensive. The purpose of this dissertation is to model salinity, detect salinity changes, and evaluate the impact of salinity in Florida Bay using remote sensing and geospatial information sciences (GIS) techniques. The specific objectives are to: 1) examine the capability of Landsat multispectral imagery for salinity modeling and monitoring; 2) detect salinity changes by building a series of salinity maps using archived Landsat images; and 3) assess the capability of spectroscopy techniques in characterizing plant stress / canopy water content (CWC) with varying salinity, sea level rise (SLR), and nutrient levels. Geographic weighted regression (GWR) models created using the first three imagery components with atmospheric and sun glint corrections proved to be more correlated (R^2 = 0.458) to salinity data versus ordinary least squares (OLS) regression models (R^2 = 0.158) and therefore GWR was the ideal regression model for continued Florida Bay salinity assessment. J. roemerianus was also examined to assess the coastal Everglades where salinity modeling is important to the water-land interface. Multivariate greenhouse studies determined the impact of nutrients to be inconsequential but increases in salinity and sea level rise both negatively affected J. roemerianus. Field spectroscopic data was then used to ascertain correlations between CWC and reflectance spectra using spectral indices and derivative analysis. It was determined that established spectral indices (max R^2 = 0.195) and continuum removal (max R^2= 0.331) were not significantly correlated to CWC but derivative analysis showed a higher correlation (R^2 = 0.515 using the first derivative at 948.5 nm). These models can be input into future imagery to predict the salinity of the South Florida water ecosystem. / Includes bibliography. / Dissertation (Ph.D.)--Florida Atlantic University, 2016. / FAU Electronic Theses and Dissertations Collection