1 |
Localisation d'objets urbains à partir de sources multiples dont des images aériennes / Localization of urban objects from multiple sources, including aerial imageryPibre, Lionel 30 November 2018 (has links)
Cette thèse aborde des problèmes liés à la localisation et la reconnaissance d’objets urbains dans des images multi-sources (optique, infrarouge, Modèle Numérique de Surface) de très haute précision acquises par voie aérienne.Les objets urbains (lampadaires, poteaux, voitures, arbres…) présentent des dimensions, des formes, des textures et des couleurs très variables. Ils peuvent être collés les uns les autres et sont de petite taille par rapport à la dimension d’une image. Ils sont présents en grand nombre mais peuvent être partiellement occultés. Tout ceci rend les objets urbains difficilement identifiables par les techniques actuelles de traitement d’images.Dans un premier temps, nous avons comparé les approches d’apprentissage classiques, composées de deux étapes - extraction de caractéristiques par le biais d’un descripteur prédéfini et utilisation d’un classifieur - aux approches d’apprentissage profond (Deep Learning), et plus précisément aux réseaux de neurones convolutionnels (CNN). Les CNN donnent de meilleurs résultats mais leurs performances ne sont pas suffisantes pour une utilisation industrielle. Nous avons donc proposé deux améliorations.Notre première contribution consiste à combiner de manière efficace les données provenant de sources différentes. Nous avons comparé une approche naïve qui consiste à considérer toutes les sources comme des composantes d’une image multidimensionnelle à une approche qui réalise la fusion des informations au sein même du CNN. Pour cela, nous avons traité les différentes informations dans des branches séparées du CNN. Nous avons ainsi montré que lorsque la base d’apprentissage contient peu de données, combiner intelligemment les sources dans une phase de pré-traitement (nous combinons l'optique et l'infrarouge pour créer une image NDVI) avant de les donner au CNN améliore les performances.Pour notre seconde contribution, nous nous sommes concentrés sur le problème des données incomplètes. Jusque-là, nous considérions que nous avions accès à toutes les sources pour chaque image mais nous pouvons aussi nous placer dans le cas où une source n’est pas disponible ou utilisable pour une image. Nous avons proposé une architecture permettant de prendre en compte toutes les données, même lorsqu’il manque une source sur une ou plusieurs images. Nous avons évalué notre architecture et montré que sur un scénario d’enrichissement, cette architecture permet d'obtenir un gain de plus de 2% sur la F-mesure.Les méthodes proposées ont été testées sur une base de données publique. Elles ont pour objectif d’être intégrées dans un logiciel de la société Berger-Levrault afin d’enrichir les bases de données géographiques et ainsi faciliter la gestion du territoire par les collectivités locales. / This thesis addresses problems related to the location and recognition of urban objects in multi-source images (optical, infrared, terrain model) of very high precision acquired by air.Urban objects (lamp posts, poles, car, tree...) have dimensions, shapes, textures and very variable colors. They can be glued to each other and are small with respect to the size of an image. They are present in large numbers but can be partially hidden. All this makes urban objects difficult to identify with current image processing techniques.First, we compared traditional learning approaches, consisting of two stages - extracting features through a predefined descriptor and using a classifier - to deep learning approaches and more precisely Convolutional Neural Networks (CNN). CNNs give better results but their performances are not sufficient for industrial use. We therefore proposed two contributions to increase performance.The first is to efficiently combine data from different sources. We compared a naive approach that considers all sources as components of a multidimensional image to an approach that merges information within CNN itself. For this, we have processed the different information in separate branches of the CNN.For our second contribution, we focused on the problem of incomplete data. Until then, we considered that we had access to all the sources for each image but we can also place ourselves in the case where a source is not available or usable. We have proposed an architecture to take into account all the data, even when a source is missing in one or more images. We evaluated our architecture and showed that on an enrichment scenario, it allows to have a gain of more than 2% on the F-measure.The proposed methods were tested on a public database. They aim to be integrated into a Berger-Levrault company software in order to enrich geographic databases and thus facilitate the management of the territory by local authorities.
|
2 |
Transformation Of An Urban Vector: Eskisehir Highway, AnkaraTekin, Tugba 01 September 2008 (has links) (PDF)
The urban transformation of the city of Ankara as a planned and constructed city with
stable configurations, definitive forms, limits and boundaries in scope of
modernization project of the country, is under the influence of new space-time
understanding with hybridizations, proximities, frictions, overlaps and
superpositions in neo-liberal era.
This thesis presents the rapid transformation of Eskisehir Highway which is the
development corridor of the city of Ankara. Eskisehir Highway is chosen in order to
reflect the complexity of the metropolitan condition of the city of Ankara. Eskisehir
Highway will be questioned as a vector which will be used as a tool to decipher
multi-dimensional dynamics of this complex urban condition which reconfigures the
new urban architecture with intensity, movement, direction and magnitude as
both the features of the vector and era.
The Highway as a vectorial urban realm is transforming itself and the nearby, with
the non-linear capitalist project production process. In order to understand this
transformation, the new urban objects of globalization will be examined as big
projects of large capital regarding the new relation patterns between architecture and
the urbanism under a framework shaped by the notions of movement-fluidity-
speed, intensity, direction and magnitude.
|
3 |
Framhävning av urbana objekt: Bakgrundsljusets påverkan på upplevelsen av ett landmärke / Emphasizing of urban objects: The influence of background light on the experience of a landmarkHallner, Ellen, Forsberg, Linnea January 2023 (has links)
Urbana miljöers komplexitet och rörelse kräver noga planerad ljussättning och omsorgsfull prioritering i samspelet mellan att lyfta fram viktiga element och lämna andra åt sidan. För att minska risken för överbelysning genom ljussättning av objekt med stor ljushetskontrast mot dess bakgrund bör alternativa metoder för att framhäva landmärken i stadsrum utredas. Detta examensarbete undersöker hur upplevelsen av mindre urbana landmärken påverkas av ljushetskontraster i dess bakgrund. Studiens syfte är att undersöka hur landmärkenas bakgrundsbelysning utformas idag, samt hur framhävning av landmärken påverkas av ljuskontrasten i dess bakgrund. Genom att använda ljussättning av landmärkenas bakgrundsmiljö som ett verktyg, syftar studien till att bidra till att skapa en tydlig och lättförståelig orientering genom stadsmiljön under mörka timmar. Studien genomfördes i tre delmoment genom ett fältexperiment, intervjuer och ett kvasi-experiment. Observation av utvalda urbana objekt och dess omgivning genomfördes i fältexperimentet. Detta följt av två semistrukturerade intervjuer kring ljusdesigners beaktningstagande av bakgrunden vid ljussättning av landmärken i urban miljö. Vid observation av studiens kvasi-experiment analyserades generella miljöer med olika bakgrundsprinciper i en iscensatt urban miljö. Resultaten av studien visar att olika stort beaktningstagande tas angående ljussättningen av urbana objekts omgivning, beroende på tidigare erfarenhet av ljusplanering. Studiens kvasi-experiment visade att ljushetskontrasten mellan ett urbant objekt och dess bakgrund, tillsammans med ljushetskontraster och formen på ljusbilderna inom bakgrunden, påverkar hur objektet uppfattas. Resultaten indikerar att en ljussättning av en urban miljö, med hänsyn till ovanstående faktorer, har påverkan på upplevelsen av objekten i miljön. Då bakgrunden är ljussatt lågmält, symmetriskt och då hela bakgrunden är definierbar ges en uppfattning om den omgivande miljön, vilket gör att objektet framhävs. / The complexity and movement of urban environments require carefully planned lighting and prioritization in the interplay between highlighting important elements and leaving others aside. In order to reduce the risk of over-lighting by lighting objects with a large brightness contrast against their background, alternative methods for highlighting landmarks in urban spaces should be investigated. This thesis examines how the experience of smaller urban landmarks is affected by brightness contrasts in its background. The purpose of the study is to investigate how the background lighting of landmarks is designed and is being designed today, as well as how highlighting of landmarks is affected by the light contrast in its background. By using lighting of the landmarks background environment as a tool, the study aims to help create a clear and easy-to-understand orientation through the urban environment during dark hours. The study was carried out in three phases through a field experiment, interviews and a final quasi-experiment. An observation of selected urban objects and their areas was carried out in the field experiment, followed by two semi-structured interviews regarding lighting designers consideration of the background when lighting landmarks in an urban environment. When observing the studys quasi-experiment, general environments with different background principles were analyzed in a staged urban environment. The results of the study show that varying degrees of consideration are taken regarding the lighting of urban objects surroundings, depending on previous experience with lighting planning. The studys quasi-experiment showed that the brightness contrast between an urban object and its background, together with brightness contrasts and the shape of the light images within the background, affects how the object is perceived. The results indicate that the lighting of an urban environment, taking these factors into account, has an impact on the experience of the objects in the environment. When the background is lit softly, symmetrically, and when the entire background is definable, an idea of the surrounding environment is given, which makes the object emphasized.
|
Page generated in 0.0433 seconds