Riparian areas are ecologically and economically critical habitats in the Canadian Prairies. An estimated 80% of riparian zones in North America are threatened by anthropogenic development. While riparian conservation is integrated into agricultural, watershed, and forestry best management practices across Canada, existing riparian health assessments are reliant on resource-intensive field surveys. The objective of this thesis was to develop a riparian health assessment using high-resolution remotely sensed imagery. Riparian health surveys were conducted along the La Salle River. High-resolution imagery and LiDAR data were integrated into an object-based image analysis of vegetation. Topographic analysis was conducted using a high-resolution DEM. These data were input into a linear discriminant classifier to model riparian health. Riparian health models containing both vegetation and topographic variables, and only vegetation variables, produced good agreement with field assessments. LiDAR data and the object-based image analysis method were successfully used to develop a remote riparian health assessment.

Made available in DSpace on 2014-06-12T16:31:35Z (GMT). No. of bitstreams: 2 arquivo7046_1.pdf: 8124499 bytes, checksum: d797c1b18a99cb9b9b46c709a4d14427 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2011 / O crescimento acelerado do processo de expansão urbana gera situações de vulnerabilidade socioambiental através do surgimento de assentamentos espontâneos, os quais são ocupados por populações de baixa renda, sem infra-estrutura nem serviços urbanos. crescimento acelerado do processo de expansão urbana gera situações de vulnerabilidade socioambiental através do surgimento de assentamentos espontâneos os quais são ocupados por populações de baixa renda e sem infra-estrutura e serviços urbanos. O presente estudo procurou identificar através de uma análise baseada em objeto (OBIA) em imagens de alta resolução identificar áreas de assentamentos subnormais resultante do crescimento do município de João Pessoa. Os bairros de Manaíra, Bairro dos Ipês e Alto do céu foram escolhidos como área de estudo por terem características urbanísticas e sociais distintas. O estudo se baseou na segmentação multiresolução e na elaboração de critérios baseados na forma, textura, cor, contexto e na relação entre super-objetos e objetos. A classificação foi estrutura em uma árvore de processo, onde documentou todo procedimento partindo da segmentação a estruturação dos critérios a associação dos objetos da imagem as classes temáticas. Possibilitando a utilização desta informação em outras imagens com característica similares. As características passíveis de serem identificadas através de imagens de sensoriamento orbital de alta resolução são referentes às propriedades urbanísticas e ambientais, que neste trabalho foram utilizadas as irregularidades de ruas, lotes não inseridos em quadras, proximidades ou invasão de áreas de proteção ambientais e leitos de rios, irregularidade tamanhos e forma das edificações. O material utilizado nesta pesquisa foi imagens do sensor QuickBird NATURAL COLOR (3 bandas multi-espectrais fusionadas com a PAN) e o software Definiens Developer 7 para segmentação e classificação. A segmentação foi estruturada em três níveis através da estratégia de classificação top-donw. As classes temáticas utilizadas foram quadras, estradas, vegetação, estradas (formal e informal), telhados vermelhos, telhados escuros, telhados/objetos claros, assentamentos subnormais e assentamentos informais. A fim de avaliar acurácia do produto gerado pela classificação foram adotados o índice de exatidão global e o coeficiente Kappa, os valores encontrados foram maiores que 80% e 0,75 respectivamente para cada método. As utilizações de informações temáticas possibilitaram uma melhor separação dos elementos asfalticos presente nas estradas e em algumas edificações, da mesma forma ocorreu entre a classe solo exposto e telhados cerâmicos. Verificou-se que os critérios estabelecidos condizem com a realidade dos assentamentos subnormais na cidade de João Pessoa. O método possibilita produzir informações sobre assentamentos subnormais coerentes com os níveis exigidos para estudos urbanos

Made available in DSpace on 2014-06-11T19:22:25Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-02-15Bitstream added on 2014-06-13T18:49:04Z : No. of bitstreams: 1 torrijoscadena_g_me_prud.pdf: 2866179 bytes, checksum: 72e4cd0bdc70c749eb5f98a106e0b51e (MD5) / A circulação de veículos na cidade de Bogotá, capital da Colômbia, é muito alta, principalmente por este ser o centro de convergência do sistema de transportes, além de ser o pólo comercial, cultural e industrial do país. Com o crescimento urbano e econômico da cidade, o número de veículos que trafega nela vem aumentando, ano após ano, principalmente na região metropolitana. Em decorrência desse aumento, está sendo observada a deterioração, cada vez maior, das vias urbanas da cidade, tornando necessário buscar alternativas que possam mitigar este problema. Neste contexto, a proposta central desta pesquisa é classificar os tipos de pavimentos das vias urbanas, de Bogotá, fazendo uso de ortoimagens fornecidas pelo Instituto Geográfico “Agustín Codazzi” da Colômbia, adquiridas com uma câmara Vexcel Ultracam-D... / The flow of vehicles in Bogotá, capital city of Colombia, is really intense, mainly for the city being the converging center of the transportation system, furthermore, being the countries’ commercial, cultural and industrial pole. Because of the urban and economic growth, the number of vehicles which are driven in that city have been increasing year after year, mainly in the metropolitan area. Due to this increase, it has been observed, each time more, the deterioration of urban ways, becoming necessary to look for alternatives which can reduce this problem. Within this context the main proposal of this research is to classify the types of pavements of Bogota’s urban ways. It’ll make use of orthoimages provided by the Geographical Institute “Agustín Codazzi” from Colombia. These picture were taken with a spacial resolution camera named Vexcel Ultracam-D ... (Complete abstract click electronic access below)

[pt] Esta dissertação visa investigar técnicas de segmentação semântica para a análise de dados de observação da Terra. Dois são os objetivos perseguidos neste estudo. O primeiro é avaliar o potencial de técnicas de segmentação semântica como opção aos métodos tradicionais de segmentação de imagens que tipicamente ignoram a informação semântica. O segundo objetivo consiste em comparar a segmentação semântica com a abordagem típica baseada em objeto (OBIA). O estudo apoia-se numa implementação de segmentação semântica baseada em Campos Aleatórios Condicionais. A estratégia baseada em objeto é representada neste estudo pelo algoritmo de segmentação conhecido como Multiresolução. O classificador Florestas Aleatórias (Random Forests) é utilizado para gerar os chamados potenciais de associação dos campos aleatórios condicionais, bem como para realizar a tarefa de classificação na cadeia de processamento típico da abordagem baseada em objeto. Experimentos realizados sobre duas imagens de altíssima resolução espacial (8 cm) indicaram uma clara superioridade da segmentação semântica, tanto em termos de acurácia espacial quanto de acurácia temática. Embora carentes de uma análise mais aprofundada que permita a generalização de suas conclusões, os resultados obtidos no presente estudo provêm elementos suficientes para encorajar a pesquisa futura sobre a aplicação da segmentação semântica na composição de estratégias sofisticadas de classificação de imagens, em particular sendo parte de modelos baseadas em objeto. / [en] This dissertation investigates semantic segmentation techniques for the analysis of Earth observation data. This study has two main task. The first one is to assess the potential of semantic segmentation techniques as an option to traditional image segmentation methods that typically ignore the semantic information. The second objective is to compare the semantic segmentation with the typical objectbased approach (OBIA). The study is based on an implementation of semantic segmentation based on Conditional Random Fields. The object-based approach is represented in this study by the segmentation algorithm known as Multiresolution. The Random Forests classifier is used to generate the association potentials for the conditional random fields and to perform the classification task in a representative implementation of the typical object-based approach. Experiments carried out on two high spatial resolution images (8 cm) indicated a clear superiority of semantic segmentation, both in terms of spatial accuracy and thematic accuracy. Although a more extensive analysis is required for the generalization of the aforementioned conclusions, the results of this study provide enough evidence to encourage a future research on the use of semantic segmentation to compose sophisticated image classification models, in particular being part of models inspired in the OBIA approach.

7 Automatic extraction of buildings and imprevious areas from very high resolution data in suburban area of Prague Abstract Nowdays, when the very high resolution satellite imagery and airbone laser scanning data have became more accessible, the possibility of their use for different types of applications increased also. With a rapid development of urban hinterlands the demand to monitor these areas increases also with the goal to avoid uncoordinated construction. This work therefore focuses on an object oriented based classification in order to design its own methodological approach for the extraction of buildings and imprevious areas in selected areas of commercial suburbanization in the Prague hinterland. The aim is among others to find the classification rules for distinguishing different types of roofs and impervious areas depending on the material, shape characteristic etc. The literature overview summarizes methods of buildings and imprevious areas classification and extraction using very high resolution optical data and elevation data. Very high resolution QuickBird imagery and airbone laser scanning LIDAR elevation data and object oriented classification methods were used for and analysis of selected commercial suburbanization model areas in Prague hinterland. The proposed methodology uses...

Submitted by Valquíria Barbieri (kikibarbi@hotmail.com) on 2018-05-10T22:30:29Z No. of bitstreams: 1 DISS_2014_Leandro Ribeiro Teixeira.pdf: 2645176 bytes, checksum: cfaa0cfd42394df366676a58482a40e4 (MD5) / Approved for entry into archive by Jordan (jordanbiblio@gmail.com) on 2018-05-24T14:20:10Z (GMT) No. of bitstreams: 1 DISS_2014_Leandro Ribeiro Teixeira.pdf: 2645176 bytes, checksum: cfaa0cfd42394df366676a58482a40e4 (MD5) / Made available in DSpace on 2018-05-24T14:20:10Z (GMT). No. of bitstreams: 1 DISS_2014_Leandro Ribeiro Teixeira.pdf: 2645176 bytes, checksum: cfaa0cfd42394df366676a58482a40e4 (MD5) Previous issue date: 2014-02-19 / Este capítulo teve como objetivo realizar uma revisão de literatura para o tema proposto neste trabalho. Desta forma, o texto apresenta a caracterização e contextualização da área de estudo, uma revisão sobre o avanço das tecnologias provenientes do Sensoriamento Remoto, técnicas de segmentação e classificação de imagens utilizando algoritmos de Classificação Orientada a Objetos, bem como o comportamento espectral da vegetação e os conceitos envolvendo espectroscopia de reflectância no âmbito das contribuições aos estudos de mapeamento do Bioma Cerrado. / This chapter aims to perform a literature review for the proposed work in this theme. Thus, the text presents the characteristics and context of the study area, a review of the advancement of technologies provided by Remote Sensing techniques, segmentation and image classification algorithms using Object - Oriented Classification and the spectral behavior of vegetation and concepts involving reflectance spectroscopy in the context of contributions to mapping studies of the Cerrado Biome.

Forest stand delineation is an essential task of forest management planning which can be time consuming and exposed to subjectivity. The increasing availability of LiDAR data and multispectral imagery offers an opportunity to improve stand delineation by means of remotely-sensed data. Under these premises, ASTER imagery and low-density LiDAR data have been used to automatically delineate forest stands in several forests of Navarra (Spain) through Object-Based Image Analysis (OBIA). Canopy cover, mean height and the canopy model have been extracted from LiDAR data and, along with VNIR ASTER bands, introduced in OBIA for forest segmentation. The outcome of segmentation has been contrasted, on the one hand, assessing segments’ inner heterogeneity. On the other, OBIA’s segments and existing stand delineations have been compared with a new method of geometrical fitting which has been ad hoc designed for this study. Results suggest that low-density LiDAR and multispectral data, along with OBIA, are a powerful tool for stand delineation. Multispectral images have a limited predicting utility for species differentiation and, in practical terms, they help to discriminate between broad-leaved, conifer and mixed stands. The performance of ASTER data, though, could be improved with higher spatial resolution VNIR imagery, specifically sub-metric VNIR orthophotos. LiDAR data, in contrast, offers a great potential for forest structure depiction. This perspective is connected with the increasingly higher resolution datasets which are to be provided by public institutions and the rapid development of drone technology. Complexity of OBIA may limit the use of this technique for small consulting firms but it is an advisable instrument for companies and institutions involved in major forestry projects. / No

In this thesis, an approach to automatically derive information about land cover from the remotely sensed data is presented. The data interpretation was done with classification process and performed in software eCognition Developer. The Object-based image analysis, which assignes the classes - for example land cover types, to clusters of pixels (=objects), was used. For the classification, products of two different data sources were combined - the orthophotos generated from aerial imagery and Normalized Digital surface model derived from LiDAR data. Five types of landscape elements were identified and classified.

L’armature verte urbaine rassemble tous les éléments de végétation contribuant à répondre aux défis majeurs associés à l’Anthropocène, comme l’adaptation au changement climatique, la préservation de la biodiversité, l’infiltration des eaux pluviales ou l’amélioration de la santé des populations. La cartographie précise des espaces végétalisés urbains fait aujourd’hui défaut bien que les données spatiales issues de capteurs aéroportés soient disponibles depuis de nombreuses années. De ce fait, il est aujourd’hui difficile d’évaluer la fonctionnalité des armatures vertes. Pour répondre à ce besoin d’évaluation, il s’est agi, dans le cadre d’étude de la Métropole de Lyon, de cartographier le territoire à un mètre de résolution spatiale entre 1984 et 2015 en utilisant une procédure orientée objet, et de rendre ces cartographies disponibles sur une plateforme web dédiée. Il en ressort que les surfaces végétalisées de la métropole sont comparables à celles d’autres agglomérations de même rang, mais que l’accès de la population à cette végétation est très inégalement réparti. Par exemple, seuls 11 % des habitants disposent d’un espace vert de plus de 2ha à moins de 5 minutes de chez eux. Souvent sous-estimés dans les politiques urbaines, les jardins des particuliers représentent deux fois la surface des espaces verts publics, et devraient faire l’objet des mêmes questionnements, quant à leur valeur pour la mitigation climatique ou pour la biodiversité. L’évaluation de l’armature urbaine ne s’arrête pas à sa cartographie, mais doit conduire à l’exploration de sa perception à différentes échelles spatiales, pour différents acteurs (habitants, professionnels et autres organismes vivants) et selon plusieurs mesures de bien-être physique, mental et culturel. / The urban green infrastructure brings together all the elements of vegetation that contribute to meeting the major challenges associated with the Anthropocene, such as adapting to climate change, preserving biodiversity, infiltrating rainwater or improving the health of populations. The precise mapping of urban green spaces is today lacking although spatial data from airborne sensors have been available for many years. As a result, it is now difficult to assess the functionality of greeninfrastructures. To meet this need for evaluation, the territory of the urban area of Lyon was mapped at one meter resolution between 1984 and 2015 using an object-oriented procedure. All maps generated were made available on a dedicated web platform. The results show that the green areas of the urban area of Lyon are comparable to those of other agglomerations of the same rank, but that the population’s access to this vegetation is very unevenly distributed. For example, only 11% of the inhabitants can find a green space of more than 2ha closer than a 5 minutes’ walk from home. Often underestimated in urban policies, private gardens make up twice the surface of public green spaces, and should be the subject of the same questions about their value for climate mitigation or the conservation of biodiversity. The evaluation of urban green infrastructures only starts with theirmapping, and should lead to the exploration of its perception at different spatial scales, for different actors (inhabitants, professionals and other living organisms) and according to several measures of physical, mental and cultural well-being.

This thesis addresses three research gaps in published literature. These are, the absence of Object Based Image Analysis (OBIA) methods for urban Land Use and Land Cover (LULC) analysis in Nigeria; the inability to use Nigeriasat-1 satellite data for urban LULC analysis and monitoring urban growth in Nigeria with Shannon’s Entropy Index. Using Abuja as a case study, this research investigated the nature of land use/land cover change (LULCC). Specific objectives were: design of an object based classification method to extract urban LULC; validate a method to extract LULC in developing countries from multiple sources of remotely sensed data; apply the method to extract LULC data; use the outputs to validate an Urban Growth Model (UGM); optimise an UGM to represent patterns and trends and through this iterative process identify and prioritise the driving forces of urban change; and finally use the outputs of the land use maps to determine if planning has controlled land use development. Landsat 7 ETM (2001), Nigeriasat-1 SLIM (2003) and SPOT 5 HRG (2006) sensor data were merged with land use cadastre in OBIA, to produce land use maps. Overall classification accuracies were 92%, 89% and 96% respectively. Post classification analysis of LULCC indicated 4.43% annual urban spread. Shannon’s Entropy index for the study period were 0.804 (2001), 0.898 (2003) and 0.930 (2006). Cellular Automata/Markov analysis was also used to predict urban growth trend of 0.89% per annum. For the first time OBIA has been used for LULC analysis in Nigeria. This research has established that Nigeriasat-1 data can contribute to urban studies using innovative OBIA methods. In addition, that Shannon’s Entropy Index can be used to understand the nature of urban growth in Nigeria. Finally, the drivers of LULCC in Abuja are similar to those of planned capital cities in other developing economies. Land use developments in Abuja can provide an insight into urban dynamics in a developing country’s capital region. OBIA, Shannon’s Entropy Index and UGM can aid urban administrators and provide information for sustainable urban planning and development.