Dissertations / Theses on the topic 'Spatial information extraction'

In my dissertation, I elaborate on the work that I have done to extract temporally-anchored spatial knowledge from text, including both intra- and inter-sentential knowledge. I also detail multiple approaches to infer spatial timeline of a person from biographies and social media. I present and analyze two strategies to annotate information regarding whether a given entity is or is not located at some location, and for how long with respect to an event. Specifically, I leverage semantic roles or syntactic dependencies to generate potential spatial knowledge and then crowdsource annotations to validate the potential knowledge. The resulting annotations indicate how long entities are or are not located somewhere, and temporally anchor this spatial information. I present an in-depth corpus analysis and experiments comparing the spatial knowledge generated by manipulating roles or dependencies. In my work, I also explore research methodologies that go beyond single sentences and extract spatio-temporal information from text. Spatial timelines refer to a chronological order of locations where a target person is or is not located. I present corpus and experiments to extract spatial timelines from Wikipedia biographies. I present my work on determining locations and the order in which they are actually visited by a person from their travel experiences. Specifically, I extract spatio-temporal graphs that capture the order (edges) of locations (nodes) visited by a person. Further, I detail my experiments that leverage both text and images to extract spatial timeline of a person from Twitter.

L’extraction d’information spatiale à partir de données textuelles est désormais un sujet de recherche important dans le domaine du Traitement Automatique du Langage Naturel (TALN). Elle répond à un besoin devenu incontournable dans la société de l’information, en particulier pour améliorer l’efficacité des systèmes de Recherche d’Information (RI) pour différentes applications (tourisme, aménagement du territoire, analyse d’opinion, etc.). De tels systèmes demandent une analyse fine des informations spatiales contenues dans les données textuelles disponibles (pages web, courriels, tweets, SMS, etc.). Cependant, la multitude et la variété de ces données ainsi que l’émergence régulière de nouvelles formes d’écriture rendent difficile l’extraction automatique d’information à partir de corpus souvent peu standards d’un point de vue lexical voire syntaxique.Afin de relever ces défis, nous proposons, dans cette thèse, des approches originales de fouille de textes permettant l’identification automatique de nouvelles variantes d’entités et relations spatiales à partir de données textuelles issues de la communication médiée. Ces approches sont fondées sur trois principales contributions qui sont cruciales pour fournir des méthodes de navigation intelligente. Notre première contribution se concentre sur la problématique de reconnaissance et d’extraction des entités spatiales à partir de corpus de messages courts (SMS, tweets) marqués par une écriture peu standard. La deuxième contribution est dédiée à l’identification de nouvelles formes/variantes de relations spatiales à partir de ces corpus spécifiques. Enfin, la troisième contribution concerne l’identification des relations sémantiques associées à l’information spatiale contenue dans les textes. Les évaluations menées sur des corpus réels, principalement en français (SMS, tweets, presse), soulignent l’intérêt de ces contributions. Ces dernières permettent d’enrichir la typologie des relations spatiales définies dans la communauté scientifique et, plus largement, de décrire finement l’information spatiale véhiculée dans les données textuelles non standards issues d’une communication médiée aujourd’hui foisonnante
The extraction of spatial information from textual data has become an important research topic in the field of Natural Language Processing (NLP). It meets a crucial need in the information society, in particular, to improve the efficiency of Information Retrieval (IR) systems for different applications (tourism, spatial planning, opinion analysis, etc.). Such systems require a detailed analysis of the spatial information contained in the available textual data (web pages, e-mails, tweets, SMS, etc.). However, the multitude and the variety of these data, as well as the regular emergence of new forms of writing, make difficult the automatic extraction of information from such corpora.To meet these challenges, we propose, in this thesis, new text mining approaches allowing the automatic identification of variants of spatial entities and relations from textual data of the mediated communication. These approaches are based on three main contributions that provide intelligent navigation methods. Our first contribution focuses on the problem of recognition and identification of spatial entities from short messages corpora (SMS, tweets) characterized by weakly standardized modes of writing. The second contribution is dedicated to the identification of new forms/variants of spatial relations from these specific corpora. Finally, the third contribution concerns the identification of the semantic relations associated withthe textual spatial information

The research accomplished in this dissertation used both mathematical and statistical techniques to extract and evaluate measures of landscape temporal dynamics and spatial structure from remotely sensed data for the purpose of mapping wildlife habitat. By coupling the landscape measures gleaned from the remotely sensed data with various sets of animal sightings and population data, effective models of habitat preference were created. Measures of temporal dynamics of vegetation greenness as measured by National Oceanographic and Atmospheric Administration's Advanced Very High Resolution Radiometer (AVHRR) satellite were used to effectively characterize and map season specific habitat of the Sonoran pronghorn antelope, as well as produce preliminary models of potential yellow-billed cuckoo habitat in Arizona. Various measures that capture different aspects of the temporal dynamics of the landscape were derived from AVHRR Normalized Difference Vegetation Index composite data using three main classes of calculations: basic statistics, standardized principal components analysis, and Fourier analysis. Pronghorn habitat models based on the AVHRR measures correspond visually and statistically to GIS-based models produced using data that represent detailed knowledge of ground-condition. Measures of temporal dynamics also revealed statistically significant correlations with annual estimates of elk population in selected Arizona Game Management Units, suggesting elk respond to regional environmental changes that can be measured using satellite data. Such relationships, once verified and established, can be used to help indirectly monitor the population. Measures of landscape spatial structure derived from IKONOS high spatial resolution (1-m) satellite data using geostatistics effectively map details of Sonoran pronghorn antelope habitat. Local estimates of the nugget, sill, and range variogram parameters calculated within 25 x 25-meter image windows describe the spatial autocorrelation of the image, permitting classification of all pixels into coherent units whose signature graphs exhibit a classic variogram shape. The variogram parameters captured in these signatures have been shown in previous studies to discriminate between different species-specific vegetation associations. The synoptic view of the landscape provided by satellite data can inform resource management efforts. The ability to characterize the spatial structure and temporal dynamics of habitat using repeatable remote sensing data allows closer monitoring of the relationship between a species and its landscape.

Face à l'explosion des nouvelles technologies (mobiles, capteurs, etc.), de grandes quantités de données localisées dans l'espace et dans le temps sont désormais disponibles. Les bases de données associées peuvent être qualifiées de bases de données spatio-temporelles car chaque donnée est décrite par une information spatiale (e.g. une ville, un quartier, une rivière, etc.) et temporelle (p. ex. la date d'un événement). Cette masse de données souvent hétérogènes et complexes génère ainsi de nouveaux besoins auxquels les méthodes d'extraction de connaissances doivent pouvoir répondre (e.g. suivre des phénomènes dans le temps et l'espace). De nombreux phénomènes avec des dynamiques complexes sont ainsi associés à des données spatio-temporelles. Par exemple, la dynamique d'une maladie infectieuse peut être décrite par les interactions entre les humains et le vecteur de transmission associé ainsi que par certains mécanismes spatio-temporels qui participent à son évolution. La modification de l'un des composants de ce système peut déclencher des variations dans les interactions entre les composants et finalement, faire évoluer le comportement global du système.Pour faire face à ces nouveaux enjeux, de nouveaux processus et méthodes doivent être développés afin d'exploiter au mieux l'ensemble des données disponibles. Tel est l'objectif de la fouille de données spatio-temporelles qui correspond à l'ensemble de techniques et méthodes qui permettent d'obtenir des connaissances utiles à partir de gros volumes de données spatio-temporelles. Cette thèse s'inscrit dans le cadre général de la fouille de données spatio-temporelles et l'extraction de motifs séquentiels. Plus précisément, deux méthodes génériques d'extraction de motifs sont proposées. La première permet d'extraire des motifs séquentiels incluant des caractéristiques spatiales. Dans la deuxième, nous proposons un nouveau type de motifs appelé "motifs spatio-séquentiels". Ce type de motifs permet d'étudier l'évolution d'un ensemble d'événements décrivant une zone et son entourage proche. Ces deux approches ont été testées sur deux jeux de données associées à des phénomènes spatio-temporels : la pollution des rivières en France et le suivi épidémiologique de la dengue en Nouvelle Calédonie. Par ailleurs, deux mesures de qualité ainsi qu'un prototype de visualisation de motifs sont été également proposés pour accompagner les experts dans la sélection des motifs d'intérêts.

In neuropsychiatry, the development of brain imaging and dedicated data analysis for personalized medicine promises to predict both the evolution of diseases and responses of treatments. The ability to estimate the time course of the disease is the first step to understand the response to potential treatments, which implies the development of methods able to capture subject-specific features in addition to the discrimination between pathological conditions. However, methods that effectively characterize the neuronal activity at the whole-brain level are still lacking, and many efforts are currently made in the fields of clinical research and neuroscience to fill this gap. The above is particularly problematic to interpret functional Magnetic Resonance Imaging (fMRI) data, which are indirectly coupled with neuronal activity because of hemodynamics, yielding much slower signals than neuronal activity. We propose a multiscale method that combines a computational whole-brain model with machine learning to solve this issue. In our approach, the model relates the neuronal activity and the fMRI signals in a mechanistic fashion, allowing for access to neuronal activity down to millisecond precision. Specifically, we use a novel methodology that allows the extraction of space-time motifs at different timescales through binned time windows. Then, we use machine learning to study which range of timescales in the modeled neuronal activity is most informative to separate the brain's dynamics during rest, distinguishing subjects, tasks, and neuropsychiatric conditions. Our multiscale computational approach is a further step to study the multiple timescales of brain dynamics and predict the dynamical interactions between brain regions. Overall, this method raises outlooks to detect biomarkers and predict responses of treatments.
En neuropsiquiatría, el desarrollo de imágenes cerebrales y el análisis de datos dedicados a la medicina personalizada prometen predecir tanto la evolución de las enfermedades como las respuestas a los tratamientos. La capacidad de estimar el curso temporal de la enfermedad es el primer paso para comprender la respuesta a posibles tratamientos, lo que implica el desarrollo de métodos capaces de capturar características específicas del sujeto, además de la discriminación entre condiciones patológicas. Sin embargo, todavía faltan métodos que caractericen eficazmente la actividad neuronal a nivel de todo el cerebro, y actualmente se están haciendo muchos esfuerzos en los campos de la investigación clínica y la neurociencia. Lo anterior es particularmente problemático para interpretar los datos funcionales de las imágenes de resonancia magnética (fMRI por sus siglas en inglés), que están acoplados indirectamente con la actividad neuronal debido a la hemodinámica, lo que produce señales mucho más lentas que la actividad neuronal. En este trabajo, proponemos un método multiescala que combina un modelo computacional de cerebro completo con aprendizaje automático para resolver este problema. En nuestro enfoque, el modelo relaciona la actividad neuronal y las señales de resonancia magnética funcional de manera mecanicista, lo que permite el acceso a la actividad neuronal con una precisión de milisegundos. Específicamente, utilizamos una nueva metodología que permite la extracción de patrones espacio-temporales en diferentes escalas temporales a través de ventanas de tiempo. Después, usamos aprendizaje automático para estudiar qué rango de escalas de tiempo en la actividad neuronal modelada es más informativo, para separar la dinámica del cerebro durante el descanso, distinguiendo sujetos, tareas y condiciones neuropsiquiátricas. Nuestro enfoque computacional multiescala es un paso más para estudiar las múltiples escalas de tiempo de la dinámica del cerebro y predecir las interacciones dinámicas entre las regiones del cerebro. En general, este método aumenta las perspectivas para detectar biomarcadores y predecir la respuesta de tratamientos.

Aujourd'hui c'est un fait, nous vivons dans un monde où les enjeux écologiques, économiques et sociétaux sont de plus en plus pressants. Au croisement des différentes lignes directrices envisagées pour répondre à ces problèmes, une vision plus précise de la mobilité humaine est un axe central et majeur, qui a des répercussions sur tous les domaines associés tels que le transport, les sciences sociales, l'urbanisme, les politiques d'aménagement, l'écologie, etc. C'est par ailleurs dans un contexte de contraintes budgétaires fortes que les principaux acteurs de la mobilité sur les territoires cherchent à rationaliser les services de transport, et les déplacements des individus. La mobilité humaine est donc un enjeu stratégique aussi bien pour les collectivités locales que pour les usagers, qu'il faut savoir observer, comprendre, et anticiper.Cette étude de la mobilité passe avant tout par une observation précise des déplacements des usagers sur les territoires. Aujourd'hui les acteurs de la mobilité se tournent principalement vers l'utilisation massive des données utilisateurs. L'utilisation simultanée de données multi-sources, multi-modales, et multi-échelles permet d'entrevoir de nombreuses possibilités, mais cette dernière présente des défis technologiques et scientifiques majeurs. Les modèles de mobilité présentés dans la littérature sont ainsi trop souvent axés sur des zones d'expérimentation limitées, en utilisant des données calibrées, etc. et leur application dans des contextes réels, et à plus large échelle est donc discutable. Nous identifions ainsi deux problématiques majeures qui permettent de répondre à ce besoin d'une meilleure connaissance de la mobilité humaine, mais également à une meilleure application de cette connaissance. La première problématique concerne l'extraction d'informations de mobilité à partir de la fusion de données hétérogènes. La seconde problématique concerne la pertinence de cette fusion dans un contexte réel, et à plus large échelle. Nous apportons différents éléments de réponses à ces problématiques dans cette thèse. Tout d'abord en présentant deux modèles de fusion de données, qui permettent une extraction d'informations pertinentes. Puis, en analysant l'application de ces deux modèles au sein du projet ANR Norm-Atis.Dans cette thèse, nous suivons finalement le développement de toute une chaine de processus. En commençant par une étude de la mobilité humaine, puis des modèles de mobilité, nous présentons deux modèles de fusion de données, et nous analysons leur pertinence dans un cas concret. Le premier modèle que nous proposons permet d'extraire 12 comportements types de mobilité. Il est basé sur un apprentissage non-supervisé de données issues de la téléphonie mobile. Nous validons nos résultats en utilisant des données officielles de l'INSEE, et nous déduisons de nos résultats, des comportements dynamiques qui ne peuvent pas être observés par les données de mobilité traditionnelles. Ce qui est une forte valeur-ajoutée de notre modèle. Le second modèle que nous proposons permet une désagrégation des flux de mobilité en six motifs de mobilité. Il se base sur un apprentissage supervisé des données issues d'enquêtes de déplacements ainsi que des données statiques de description du sursol. Ce modèle est appliqué par la suite aux données agrégés au sein du projet Norm-Atis. Les temps de calculs sont suffisamment performants pour permettre une application de ce modèle dans un contexte temps-réel
Today it is a fact that we live in a world where ecological, economic and societal issues are increasingly pressing. At the crossroads of the various guidelines envisaged to address these problems, a more accurate vision of human mobility is a central and major axis, which has repercussions on all related fields such as transport, social sciences, urban planning, management policies, ecology, etc. It is also in the context of strong budgetary constraints that the main actors of mobility on the territories seek to rationalize the transport services and the movements of individuals. Human mobility is therefore a strategic challenge both for local communities and for users, which must be observed, understood and anticipated.This study of mobility is based above all on a precise observation of the movements of users on the territories. Nowadays mobility operators are mainly focusing on the massive use of user data. The simultaneous use of multi-source, multi-modal, and multi-scale data opens many possibilities, but the latter presents major technological and scientific challenges. The mobility models presented in the literature are too often focused on limited experimental areas, using calibrated data, etc., and their application in real contexts and on a larger scale is therefore questionable. We thus identify two major issues that enable us to meet this need for a better knowledge of human mobility, but also to a better application of this knowledge. The first issue concerns the extraction of mobility information from heterogeneous data fusion. The second problem concerns the relevance of this fusion in a real context, and on a larger scale. These issues are addressed in this dissertation: the first, through two data fusion models that allow the extraction of mobility information, the second through the application of these fusion models within the ANR Norm-Atis project.In this thesis, we finally follow the development of a whole chain of processes. Starting with a study of human mobility, and then mobility models, we present two data fusion models, and we analyze their relevance in a concrete case. The first model we propose allows to extract 12 types of mobility behaviors. It is based on an unsupervised learning of mobile phone data. We validate our results using official data from the INSEE, and we infer from our results, dynamic behaviors that can not be observed through traditional mobility data. This is a strong added-value of our model. The second model operates a mobility flows decompositoin into six mobility purposes. It is based on a supervised learning of mobility surveys data and static data from the land use. This model is then applied to the aggregated data within the Norm-Atis project. The computing times are sufficiently powerful to allow an application of this model in a real-time context

The video databases and retrieval of data from these databases have become popular in various business areas of work with the improvements in technology. As a kind of video database, video archive systems need user-friendly interfaces to retrieve video frames. In this thesis, an NLP based user interface to a video database system is developed using a content-based spatio-temporal video data model. The data model is focused on the semantic content which includes objects, activities, and spatial properties of objects. Spatio-temporal relationships between video objects and also trajectories of moving objects can be queried with this data model. In this video database system, NL interface enables flexible querying. The queries, which are given as English sentences, are parsed using Link Parser. Not only exact matches but similar objects and activities are also returned from the database with the help of the conceptual ontology module to return all related frames to the user. This module is implemented using a distance-based method of semantic similarity search on the semantic domain-independent ontology, WordNet. The semantic representations of the given queries are extracted from their syntactic structures using information extraction techniques. The extracted semantic representations are used to call the related parts of the underlying spatio-temporal video data model to calculate the results of the queries.

Orientador: Fernando José Von Zuben
Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação
Made available in DSpace on 2018-08-27T03:14:35Z (GMT). No. of bitstreams: 1 Kurka_DavidBurth_M.pdf: 1660677 bytes, checksum: 7258daf8129b4dac9d1f647195775d3c (MD5) Previous issue date: 2015
Resumo: Com o surgimento e a popularização de Redes Sociais Online e de Serviços de Redes Sociais, pesquisadores da área de computação têm encontrado um campo fértil para o desenvolvimento de trabalhos com grande volume de dados, modelos envolvendo múltiplos agentes e dinâmicas espaço-temporais. Entretanto, mesmo com significativo elenco de pesquisas já publicadas no assunto, ainda existem aspectos das redes sociais cuja explicação é incipiente. Visando o aprofundamento do conhecimento da área, este trabalho investiga fenômenos de compartilhamento coletivo na rede, que caracterizam eventos de difusão de informação. A partir da observação de dados reais oriundos do serviço online Twitter, tais eventos são modelados, caracterizados e analisados. Com o uso de técnicas de aprendizado de máquina, são encontrados padrões nos processos espaço-temporais da rede, tornando possível a construção de classificadores de mensagens baseados em comportamento e a caracterização de comportamentos individuais, a partir de conexões sociais
Abstract: With the advent and popularization of Online Social Networks and Social Networking Services, computer science researchers have found fertile field for the development of studies using large volumes of data, multiple agents models and spatio-temporal dynamics. However, even with a significant amount of published research on the subject, there are still aspects of social networks whose explanation is incipient. In order to deepen the knowledge of the area, this work investigates phenomena of collective sharing on the network, characterizing information diffusion events. From the observation of real data obtained from the online service Twitter, we collect, model and characterize such events. Finally, using machine learning and computational data analysis, patterns are found on the network's spatio-temporal processes, making it possible to classify a message's topic from users behaviour and the characterization of individual behaviour, from social connections
Mestrado
Engenharia de Computação
Mestre em Engenharia Elétrica

Les systèmes de géolocalisation permettent la surveillance en temps réel des déplacements d'objets mobiles. Aujourd'hui, les données produites par ces capteurs sont reçues et stockées dans des bases de données spatio-temporelles. Un processus de fouille de données appliqué sur ces bases de données spatio-temporelles permet d'extraire le comportement des objets mobiles (patrons spatio-temporels) et d'analyser en temps réel les trajectoires d'objets mobiles suivant un même itinéraire. En utilisant ces modèles, des situations inhabituelles peuvent être détectés. Cette thèse définit à la fois des patrons spatio-temporels ainsi que des outils de comparaison et de qualification de trajectoires en utilisant un indice de similarité basée sur des mesures spatiales et temporelles et la logique floue. Ces outils peuvent être utilisés pour faciliter la surveillance du trafic maritime.

博士
國立中興大學
水土保持學系
90
Watershed unit has been regarded as an analyzed object for slopeland disaster assessment. With the fast growing progress of computer technologies, instead of manual operation, there is a trend of applying Digital Elevation Model (DEM) in automated watershed delineating and drainage network design. This study first focused on improving extraction method for automated watershed delineating. The concepts for depression outlets decision, looped drainage direction calculation, reasonable watersheds area delineation, and simulation of realistic drainage networks are all revised to propose a new approach for more reasonable and efficient in watershed delineation. Three major reservoir watersheds (Shihmen, Techi and Tsenwen) were chosen to calculate watershed geomorphologic and hydrologic information for verifying the suitability of proposed approach. An expert system was also developed using the approach included Universal Soil Loss Equation (USLE) and Sediment Delivery Ratio (SDR) to estimate watershed sediment yield. The system shows good performance for Shihmen, Techi, Tsenwen reservoir and Erzen creek watersheds. The results are summarized as follows: Using DEM to derive drainage directions of a watershed is frequently used in recent study. However, determinations of the optimal outlet and drainage directions for the depressions should be improved for fitting the real field data. This study proposed surface-inclining approach to couple with elevation-differencing approach (Jenson and Domingue, 1988) for determining incipient drainage directions. The calculation of optimal outlet and drainage direction in the depressions can be performed using watershed depression approach with PROMETHEE theory. Compared with elevation-smoothing, depression-filling and repeatedly elevation-incrementing approaches, the drainage directions for Shihmen, Techi and Twensen reservoir watersheds calculated by watershed depression approach show more reasonable and realistic outcomes. A dynamic extraction technique for tracing upstream drainage area based on user-specified outlet is also developed for fast automatically watershed delineating. The watershed area extracted for Shihmen, Techi and Twensen reservoir with the outlet located at the site of dam is 75634 ha, 52367 ha and 48396 ha, respectively. Due to inhomogeneous characteristics of geomorphology, geology, soil and/or climates of a watershed, the real drainage networks can not be delineated properly using single-threshold approach. Instead, a multiple-threshold approach is developed to cope with the real spatial distribution of streams in a watershed. Annual watershed sediment yield calculated for Shihmen, Techi, Twensen reservoirs and Erzen creek is 1928168 tons/yr, 1793742 tons/yr, 4807205 tons/yr and 5074396 tons/yr, and the corresponding annual erosion depth is 1.82mm, 2.45mm, 7.10mm and 25.81mm, respectively. Tsenwen reservoir and Erzen creek show higher erosion depth because of moderate mudstone distribution in the watershed. Model efficiency by Nash and Sutcliffe (1970) for sediment yield estimation is 81.21%, the model ( , ) shows significant correlation between estimated ( ) and measured ( ) data. The sites with can be clustered as higher sediment delivery sites from characteristic curve analysis. Area percentage of higher sediment delivery sites occupied for Shihmen, Techi and Tsenwen reservoir watersheds are only 10.58%, 9.21% and 10.01%. While the sediment yield percentage for the respective watershed can reach to 46.99%, 49.20% and 46.47%. It shows that higher sediment delivery sites are the main source of soil erosion.

Historical maps are fascinating documents and a valuable source of information for scientists of various disciplines. Many of these maps are available as scanned bitmap images, but in order to make them searchable in useful ways, a structured representation of the contained information is desirable. This book deals with the extraction of spatial information from historical maps. This cannot be expected to be solved fully automatically (since it involves difficult semantics), but is also too tedious to be done manually at scale. The methodology used in this book combines the strengths of both computers and humans: it describes efficient algorithms to largely automate information extraction tasks and pairs these algorithms with smart user interactions to handle what is not understood by the algorithm. The effectiveness of this approach is shown for various kinds of spatial documents from the 16th to the early 20th century
Historische Landkarten sind faszinierende Dokumente und eine wertvolle Informationsquelle für Wissenschaftler verschiedener Fächer. Viele dieser Karten liegen als gescannte Bitmap-Bilder vor, aber um sie auf nützliche Art durchsuchbar zu machen ist eine strukturierte Repräsentation der enthaltenen Informationen wünschenswert. Dieses Buch beschäftigt sich mit der Extraktion räumlicher Informationen aus historischen Landkarten. Man kann nicht erwarten, dass dies vollautomatisch geschieht (da komplizierte Semantik involviert ist), aber es ist auch zu aufwändig, um im großen Stil manuell durchgeführt zu werden. Die Methodik, die in diesem Buch verwendet wird, kombiniert die Stärken von Computern und Menschen: Es werden effiziente Algorithmen beschrieben, die Extraktionsaufgaben weitgehend automatisieren, und dazu passende Nutzerinteraktionen entworfen, mit denen Fälle gelöst werden, die die Algorithmen nicht vestehen. Die Effekitivität dieses Ansatzes wird anhand verschiedener räumlicher Dokumente aus dem 16. bis frühen 20. Jahrhundert gezeigt

博士
國立成功大學
測量及空間資訊學系碩博士班
93
3D city model or 3D spatial information system integrates 3D spatial information and thus can be applied to versatile applications such as urban planning, real estate managing, traffic evaluating, radio station locating, tourist guiding etc. Photogrammetry has been considered as the most efficient technique for extracting 3D spatial information in the last few decades. However, the operator has to measure the 3D coordinates of all points and connect the adjacent points in sequence. This point-by-point measurement of using floating mark has become the bottleneck while reconstructing the 3D city model. In this paper, we expanded the floating mark to the floating model based on the concept of model-based building extraction. The measuring tool is no longer only an abstract point but also many kinds of 3D model, which can be scaled, rotated, or moved in the space. The floating model is defined with a datum point whose 3D coordinates indicate the spatial position of the model as the floating mark does. Furthermore, each kind of models is associated with a set of pose parameters to describe its rotation about the three orthogonal axes and shape parameters to describe its scales along predefined directions. In other words, the floating model is a flexible entity floating in the space, and can be adjusted to fit the object by these parameters. If the model parameters are good enough to represent the 3D spatial information of the object, the projection of the floating model on every overlapped image will all be coincident to the object's image. Therefore, the basic idea of the floating model theory is to fit model to the overlapped images by adjusting the model parameters. In order to meet the needs of building reconstruction, we designed several primitive models and established a model-base that is composed of four kinds of models: point, line segment, surface, and solid. By means of human-computer interface, the operator is able to choose the most suitable model and measure the object on multiple images simultaneously. For the building with complicated structure, the extraction can be implemented following the Constructive Solid Geometry(CSG) according to the required level of details. The building can be split into several parts, modeled part-by-part, and finally joined into a complex model by Boolean Set Operators. Based on the floating model theory and model-based building extraction, we proposed a semi-automated extraction strategy. A friendly human-machine interface is designed for the operator to choose and adjust the floating model to fit the images manually. Then, the computer calculate the optimal fit by an ad hoc Least-Squares Model-Image Fitting(LSMIF) algorithm. Thus the 3D spatial information can be extracted object-by-object rather than point-by-point by means of floating model, which increases the efficiency and accuracy. Tested by a series of experiments on real aerial images, the proposed floating model has shown its capability and potential for 3D spatial information extraction.

碩士
國立臺灣大學
地理環境資源學研究所
91
Along with the network development, the electronics document presents the growth of the geometric progression, among them, the news document is everyday constantly at produce, accumulate a great deal of information. If can accomplish each news to event topic, relational person, taking place time, occurrence location etc, according to the attribute, then it can make these information more application, for example: The query, management of the data, and the application of the relational database etc… The application of the spatial information is a lot of, have something to do with the location geocoding, map display, for example the addresses or roads query on the map, the map display of governments'' social-economic data, such as crime location distribution at each district, line distribution etc. If these maps connect with other databases, it can further make the spatial decision support. If we can extract the hidden spatial information in the news document, then it can manage a great deal of news document effectively, and uses the information quickly. The types of geographic location representation are a lot of. They can be classified into point, line, and polygon kinds of geographic location representation. This research emphasize at point and a few line kinds description. This research with the way that inducing the news document, making use of the Dictionary Method and Linguistic Method in the natural language approach, through Information Map system, construct the extracting rules an a system for the news documents, with the fire, rob, the traffic accident topics. The information in topics include topic key word, taking place time and occurrence location. The rule to extract event occurrence location can be classify to five types: Address, Non-urban Road, Intersection Road, Mark, Dynamic Intersection Road Location Description; The rule to extract time description can be classified into six types: Season, Date, Week, Quarter, Festival and Time. Joining with the simple event keyword extracting rules, regards the paragraph as the unit to pick, the extraction export structure database format, to provide the user to make more exploitations. The system test shows that these methods can extract news events information effectively and correctly.

碩士
義守大學
資訊管理學系碩士班
95
Modern techniques of satellite image acquisition have been of great advance lately, which provide a great amount of images with a higher resolution both in spatial and spectral resolution. However, the rate of utilizing the existed images has not yet been sufficient in comparison to the rate of obtaining them. Hence, issues in using automated method of linear feature extraction for replacing manual process have drawn a great deal of attentions in this area lately. The purpose of this study is to develop an integrated method for extracting linear features of oceanic internal waves from satellite imagery using spatial information techniques, which include: wavelet transform based de-noise, Multiscale Retinex (MSR), and linear feature extraction (LEF). To evaluate the performance of the integrated method, the extracted linear features will be vectorized and overlapped with the original image in the Geographic Information System (GIS) to investigate the position discrepancy between them and the true features’ boundary. The results show that the MSR method provides enhanced image with improved color contrast and brightness, which result in a better quality of extracted linear features. Finally, we evaluate the performance of feature extraction using both the Canny method and the Wavelet Transform Modulus Maxima (WTMM) method. It is shown that the Canny method is superior to the WTMM method in terms of visualization quality and positioning accuracy.

Actualmente, os municípios são obrigados a produzir, no âmbito da elaboração dos instrumentos de gestão territorial, cartografia homologada pela autoridade nacional. O Plano Director Municipal (PDM) tem um período de vigência de 10 anos. Porém, no que diz respeito à cartografia para estes planos, principalmente em municípios onde a pressão urbanística é elevada, esta periodicidade não é compatível com a dinâmica de alteração de uso do solo. Emerge assim, a necessidade de um processo de produção mais eficaz, que permita a obtenção de uma nova cartografia de base e temática mais frequentemente. Em Portugal recorre-se à fotografia aérea como informação de base para a produção de cartografia de grande escala. Por um lado, embora este suporte de informação resulte em mapas bastante rigorosos e detalhados, a sua produção têm custos muito elevados e consomem muito tempo. As imagens de satélite de muito alta-resolução espacial podem constituir uma alternativa, mas sem substituir as fotografias aéreas na produção de cartografia temática, a grande escala. O tema da tese trata assim da satisfação das necessidades municipais em informação geográfica actualizada. Para melhor conhecer o valor e utilidade desta informação, realizou-se um inquérito aos municípios Portugueses. Este passo foi essencial para avaliar a pertinência e a utilidade da introdução de imagens de satélite de muito alta-resolução espacial na cadeia de procedimentos de actualização de alguns temas, quer na cartografia de base quer na cartografia temática. A abordagem proposta para solução do problema identificado baseia-se no uso de imagens de satélite e outros dados digitais em ambiente de Sistemas de Informação Geográfica. A experimentação teve como objectivo a extracção automática de elementos de interesse municipal a partir de imagens de muito alta-resolução espacial (fotografias aéreas ortorectificadas, imagem QuickBird, e imagem IKONOS), bem como de dados altimétricos (dados LiDAR). Avaliaram-se as potencialidades da informação geográfica extraídas das imagens para fins cartográficos e analíticos. Desenvolveram-se quatro casos de estudo que reflectem diferentes usos para os dados geográficos a nível municipal, e que traduzem aplicações com exigências diferentes. No primeiro caso de estudo, propõe-se uma metodologia para actualização periódica de cartografia a grande escala, que faz uso de fotografias aéreas vi ortorectificadas na área da Alta de Lisboa. Esta é uma aplicação quantitativa onde as qualidades posicionais e geométricas dos elementos extraídos são mais exigentes. No segundo caso de estudo, criou-se um sistema de alarme para áreas potencialmente alteradas, com recurso a uma imagem QuickBird e dados LiDAR, no Bairro da Madre de Deus, com objectivo de auxiliar a actualização de cartografia de grande escala. No terceiro caso de estudo avaliou-se o potencial solar de topos de edifícios nas Avenidas Novas, com recurso a dados LiDAR. No quarto caso de estudo, propõe-se uma série de indicadores municipais de monitorização territorial, obtidos pelo processamento de uma imagem IKONOS que cobre toda a área do concelho de Lisboa. Esta é uma aplicação com fins analíticos onde a qualidade temática da extracção é mais relevante.
Currently, the Portuguese municipalities are required to produce homologated cartography, under the Territorial Management Instruments framework. The Municipal Master Plan (PDM) has to be revised every 10 years, as well as the topographic and thematic maps that describe the municipal territory. However, this period is inadequate for representing counties where urban pressure is high, and where the changes in the land use are very dynamic. Consequently, emerges the need for a more efficient mapping process, allowing obtaining recent geographic information more often. Several countries, including Portugal, continue to use aerial photography for large-scale mapping. Although this data enables highly accurate maps, its acquisition and visual interpretation are very costly and time consuming. Very-High Resolution (VHR) satellite imagery can be an alternative data source, without replacing the aerial images, for producing large-scale thematic cartography. The focus of the thesis is the demand for updated geographic information in the land planning process. To better understand the value and usefulness of this information, a survey of all Portuguese municipalities was carried out. This step was essential for assessing the relevance and usefulness of the introduction of VHR satellite imagery in the chain of procedures for updating land information. The proposed methodology is based on the use of VHR satellite imagery, and other digital data, in a Geographic Information Systems (GIS) environment. Different algorithms for feature extraction that take into account the variation in texture, color and shape of objects in the image, were tested. The trials aimed for automatic extraction of features of municipal interest, based on aerial and satellite high-resolution (orthophotos, QuickBird and IKONOS imagery) as well as elevation data (altimetric information and LiDAR data). To evaluate the potential of geographic information extracted from VHR images, two areas of application were identified: mapping and analytical purposes. Four case studies that reflect different uses of geographic data at the municipal level, with different accuracy requirements, were considered. The first case study presents a methodology for periodic updating of large-scale maps based on orthophotos, in the area of Alta de Lisboa. This is a situation where the positional and geometric accuracy of the extracted information are more demanding, since technical mapping standards must be complied. In the second case study, an alarm system that indicates the location of potential changes in building areas, using a QuickBird image and LiDAR data, was developed for the area of Bairro da Madre de Deus. The goal of the system is to assist the updating of large scale mapping, providing a layer that can be used by the municipal technicians as the basis for manual editing. In the third case study, the analysis of the most suitable roof-tops for installing solar systems, using LiDAR data, was performed in the area of Avenidas Novas. A set of urban environment indicators obtained from VHR imagery is presented. The concept is demonstrated for the entire city of Lisbon, through IKONOS imagery processing. In this analytical application, the positional quality issue of extraction is less relevant.
GEOSAT – Methodologies to extract large scale GEOgraphical information from very high resolution SATellite images (PTDC/GEO/64826/2006), e-GEO – Centro de Estudos de Geografia e Planeamento Regional, da Faculdade de Ciências Sociais e Humanas, no quadro do Grupo de Investigação Modelação Geográfica, Cidades e Ordenamento do Território

碩士
大同大學
資訊工程學系(所)
94
In this thesis, the method to extract the spatial information and single representing source of original sound field in stereo, and then synthesis them as demanded are proposed. The objective is to synthesize appropriate sound field corresponding to vary listening condition. The discussed situation is focused on multi-sources playing the same melody by the same music instrument aligned in line. Since each source plays the same melody, the same music scale would be played on the sector in time. Human perception is insensitive to the phase of audio. So we might assume that the magnitudes of spectrogram of each source is similar even their waveforms are different. Therefore, the signal received by microphone could be treated as the summation of one spectrogram with shifts in time and attenuation. It is similar to an image corrupted by a motion blur function. Thus, the concept of image-restoration may be applied to extract the spatial information and single representing source by which the property of time-frequency components of each original source could be represented. The sound field similar to original sound field can be synthesized using the extracted single representing source and the obtained spatial information. Also the spatial information can be modified to synthesize the different sound field for different playback conditions in pleasure. The simulation is performed to confirm the method in this thesis. And the result shows that the concept of image distortion/restoration process with sound spectrogram could be applied to the spatial information extraction and sound field resynthesis. There will be certain compression effects with applying the concept of decomposing and re-synthesizing in this thesis with multi-channel processing in the future.

碩士
大同大學
資訊工程學系(所)
96
Recently, microphone array technique has already been used in teleconference and distance education. Accurate localization of multiple sound sources is indispensable for sound capture of microphone array. The sound signal from the position selected could also be enhanced. In this thesis, the method to extract the spatial information and sound sources from the received signal is proposed. The Cross-power Spectrum Phase (CSP) is used to nd the arrival time delay from sources to microphone. If two microphones are considered to locate at two foci, then the trajectory of the time delay found by microphone pair is a hyperbola. Treating a source location is a kind of projection. All of hyperbolas produced by microphone pairs have been collected. Therefore, the back-projection method could be used to reconstruct the source location. After all of the source locations are known, the sound projection is used to extract the sources one by one according the the spatial information. Finally, a sound eld would be synthesized. The result of the simulation veri es that the synthesized sound eld similar to original ones is synthesized. Furthermore, a new sound eld could also be synthesized according to the spatial information with requirement.

碩士
逢甲大學
土地管理研究所
85
The phrase "land inbreeds fortune" reveals not only the close relationship between land and human life, but also one of the four important factors of productivity. Land characteristics such as supply substantiality, non-mobile objects, non-alternative targets and incomplete market information, compose the differences from other commercial products. Optimum resource allocation of land can hardly be reached by free market mechanism. It is obviously that social problem coming from unequal social resources allocation and inefficiencies will be derived from non-controllable market reasonable price. There are abundant literature studies on the topic of land appraisal. The goal of this paper is not to assess land price or examine land appraisal parameter, but to construct the spatial information extracting model for factors which influence land price.The appraisers used to collect a lot of appraisal related data and conduct some onsite survey, in order to accordingly estimate the land value. Therefore workload was very heavy and time consuming. This study took the advantages of the fast-computing and mass data-storing capabilities of a computer, and mean while applied the GIS technique to the establishment of data file for the appraisal process. So that the time and labors can be saved and the efficiency of the assessment of land parcels can be enhanced. Taichung city in Taiwan was selected for an empirical test area. An existing land appraisal model is applied in this test to analyze the issues of spatial data including (1) establishment of criteria for data extraction, (2)investigation of spatial relationships among objects, such as distance, direction, and topological relationships, and (3)the design of spatial data analysis process for land appraisal.The result indicated that all spatial analysis is based on the feature topology, and network analysis is the most important tool for spatial analysis. Auto-extracting information can serve as a useful tool the assessment of external environment condition and can improve the appraisal technique. But it is non-efficient in the assessment of internal basic site information.

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.
A novel approach was used with data from the Moderate Resolution Imaging Spectroradiometer (MODIS) to detect Marine Mucilage in tow different marine areas of the Italy (Campanian Seas and North Adriatic Sea) from 2010 to 2012. The approach involves first deriving a Mucilage Index (MI) based on the medium‐resolution (500 m) MODIS reflectance data correction of the ozone/gaseous absorption and Rayleigh scattering effects and then objectively determining the MI threshold value (0.05