Tesi sul tema "Scanner mobile"

Aktuelle 3D-Stadtmodelle werden immer wichtiger in verschiedenen städtischen Anwendungsbereichen. Im Moment dienen sie als Grundlage bei der Stadtplanung, virtuellem Tourismus und Navigationssystemen. Mittlerweile ist der Bedarf an 3D-Gebäudemodellen dramatisch gestiegen. Der Grund dafür sind hauptsächlich Navigationssysteme und Onlinedienste wie Google Earth. Die Mehrheit der Untersuchungen zur Rekonstruktion von Gebäudemodellen von Luftaufnahmen konzentriert sich ausschließlich auf Dachmodellierung. Jedoch treiben Anwendungen wie Virtuelle Realität und Navigationssysteme die Nachfrage nach detaillieren Gebäudemodellen, die nicht nur die geometrischen Aspekte sondern auch semantische Informationen beinhalten, stark an. Urbanisierung und Industrialisierung beeinflussen das Wachstum von urbaner Vegetation drastisch, welche als ein wesentlicher Teil des Lebensraums angesehen wird. Aus diesem Grund werden Aufgaben wie der Ökosystemüberwachung, der Verbesserung der Planung und des Managements von urbanen Regionen immer mehr Aufmerksamkeit geschenkt. Gleichermaßen hat die Erkennung und Modellierung von Bäumen im Stadtgebiet sowie die kontinuierliche Überprüfung ihrer Inventurparameter an Bedeutung gewonnen. Die steigende Nachfrage nach 3D-Gebäudemodellen, welche durch Fassadeninformation ergänzt wurden, und Informationen über einzelne Bäume im städtischen Raum erfordern effiziente Extraktions- und Rekonstruktionstechniken, die hochgradig automatisiert sind. In diesem Zusammenhang ist das Wissen über die geometrische Form jedes Objektteils ein wichtiger Aspekt. Heutzutage, wird das Mobile Laser Scanning (MLS) vermehrt eingesetzt um Objekte im städtischen Umfeld zu erfassen und es entwickelt sich zur Hauptquelle von Daten für die Modellierung von urbanen Objekten. Eine Vielzahl von Objekten wurde schon mit Daten von MLS rekonstruiert. Außerdem wurden bereits viele Methoden für die Verarbeitung von MLS-Daten mit dem Ziel urbane Objekte zu erkennen und zu rekonstruieren vorgeschlagen. Die 3D-Punkwolke einer städtischen Szene stellt eine große Menge von Messungen dar, die viele Objekte von verschiedener Größe umfasst, komplexe und unvollständige Strukturen sowie Löcher (Rauschen und Datenlücken) enthält und eine inhomogene Punktverteilung aufweist. Aus diesem Grund ist die Verarbeitung von MLS-Punktwolken im Hinblick auf die Extrahierung und Modellierung von wesentlichen und charakteristischen Fassadenstrukturen sowie Bäumen von großer Bedeutung. In der Arbeit werden zwei neue Methoden für die Rekonstruktion von Gebäudefassaden und die Extraktion von Bäumen aus MLS-Punktwolken vorgestellt, sowie ihre Anwendbarkeit in der städtischen Umgebung analysiert. Die erste Methode zielt auf die Rekonstruktion von Gebäudefassaden mit expliziter semantischer Information, wie beispielsweise Fenster, Türen, und Balkone. Die Rekonstruktion läuft vollautomatisch ab. Zu diesem Zweck werden einige Algorithmen vorgestellt, die auf dem Vorwissen über die geometrische Form und das Arrangement von Fassadenmerkmalen beruhen. Die initiale Klassifikation, mit welcher die Punkte in Objektpunkte und Bodenpunkte unterschieden werden, wird über eine lokale Höhenhistogrammanalyse zusammen mit einer planaren Region-Growing-Methode erzielt. Die Punkte, die als zugehörig zu Objekten klassifiziert werden, werden anschließend in Ebenen segmentiert, welche als Basiselemente der Merkmalserkennung angesehen werden können. Information über die Gebäudestruktur kann in Form von Regeln und Bedingungen erfasst werden, welche die wesentlichen Steuerelemente bei der Erkennung der Fassadenmerkmale und der Rekonstruktion des geometrischen Modells darstellen. Um Merkmale wie Fenster oder Türen zu erkennen, die sich an der Gebäudewand befinden, wurde eine löcherbasierte Methode implementiert. Einige Löcher, die durch Verdeckungen entstanden sind, können anschließend durch einen neuen regelbasierten Algorithmus eliminiert werden. Außenlinien der Merkmalsränder werden durch ein Polygon verbunden, welches das geometrische Modell repräsentiert, indem eine Methode angewendet wird, die auf geometrischen Primitiven basiert. Dabei werden die topologischen Relationen unter Beachtung des Vorwissens über die primitiven Formen analysiert. Mögliche Außenlinien können von den Kantenpunkten bestimmt werden, welche mit einer winkelbasierten Methode detektiert werden können. Wiederkehrende Muster und Ähnlichkeiten werden ausgenutzt um geometrische und topologische Ungenauigkeiten des rekonstruierten Modells zu korrigieren. Neben der Entwicklung des Schemas zur Rekonstruktion des 3D-Fassadenmodells, sind die Segmentierung einzelner Bäume und die Ableitung von Attributen der städtischen Bäume im Fokus der Untersuchung. Die zweite Methode zielt auf die Extraktion von individuellen Bäumen aus den Restpunktwolken. Vorwissen über Bäume, welches speziell auf urbane Regionen zugeschnitten ist, wird im Extraktionsprozess verwendet. Der formbasierte Ansatz zur Extraktion von Einzelbäumen besteht aus einer Reihe von Schritten. In jedem Schritt werden Objekte in Abhängigkeit ihrer geometrischen Merkmale gefunden. Stämme werden unter Ausnutzung der Hauptrichtung der Punktverteilung identifiziert. Dafür werden Punktsegmente gesucht, die einen Teil des Baumstamms repräsentieren. Das Ergebnis des Algorithmus sind segmentierte Bäume, welche genutzt werden können um genaue Informationen über die Größe und Position jedes einzelnen Baumes abzuleiten. Einige Beispiele der Ergebnisse werden in der Arbeit angeführt. Die Zuverlässigkeit der Algorithmen und der Methoden im Allgemeinen wurden unter Verwendung von drei Datensätzen, die mit verschiedenen Laserscannersystemen aufgenommen wurden, verifiziert. Die Untersuchung zeigt auch das Potential sowie die Einschränkungen der entwickelten Methoden wenn sie auf verschiedenen Datensätzen angewendet werden. Die Ergebnisse beider Methoden wurden quantitativ bewertet unter Verwendung einer Menge von Maßen, die die Qualität der Fassadenrekonstruktion und Baumextraktion betreffen wie Vollständigkeit und Genauigkeit. Die Genauigkeit der Fassadenrekonstruktion, der Baumstammdetektion, der Erfassung von Baumkronen, sowie ihre Einschränkungen werden diskutiert. Die Ergebnisse zeigen, dass MLS-Punktwolken geeignet sind um städtische Objekte detailreich zu dokumentieren und dass mit automatischen Rekonstruktionsmethoden genaue Messungen der wichtigsten Attribute der Objekte, wie Fensterhöhe und -breite, Flächen, Stammdurchmesser, Baumhöhe und Kronenfläche, erzielt werden können. Der gesamte Ansatz ist geeignet für die Rekonstruktion von Gebäudefassaden und für die korrekte Extraktion von Bäumen sowie ihre Unterscheidung zu anderen urbanen Objekten wie zum Beispiel Straßenschilder oder Leitpfosten. Aus diesem Grund sind die beiden Methoden angemessen um Daten von heterogener Qualität zu verarbeiten. Des Weiteren bieten sie flexible Frameworks für das viele Erweiterungen vorstellbar sind
Up-to-date 3D urban models are becoming increasingly important in various urban application areas, such as urban planning, virtual tourism, and navigation systems. Many of these applications often demand the modelling of 3D buildings, enriched with façade information, and also single trees among other urban objects. Nowadays, Mobile Laser Scanning (MLS) technique is being progressively used to capture objects in urban settings, thus becoming a leading data source for the modelling of these two urban objects. The 3D point clouds of urban scenes consist of large amounts of data representing numerous objects with significant size variability, complex and incomplete structures, and holes (noise and data gaps) or variable point densities. For this reason, novel strategies on processing of mobile laser scanning point clouds, in terms of the extraction and modelling of salient façade structures and trees, are of vital importance. The present study proposes two new methods for the reconstruction of building façades and the extraction of trees from MLS point clouds. The first method aims at the reconstruction of building façades with explicit semantic information such as windows, doors and balconies. It runs automatically during all processing steps. For this purpose, several algorithms are introduced based on the general knowledge on the geometric shape and structural arrangement of façade features. The initial classification has been performed using a local height histogram analysis together with a planar growing method, which allows for classifying points as object and ground points. The point cloud that has been labelled as object points is segmented into planar surfaces that could be regarded as the main entity in the feature recognition process. Knowledge of the building structure is used to define rules and constraints, which provide essential guidance for recognizing façade features and reconstructing their geometric models. In order to recognise features on a wall such as windows and doors, a hole-based method is implemented. Some holes that resulted from occlusion could subsequently be eliminated by means of a new rule-based algorithm. Boundary segments of a feature are connected into a polygon representing the geometric model by introducing a primitive shape based method, in which topological relations are analysed taking into account the prior knowledge about the primitive shapes. Possible outlines are determined from the edge points detected from the angle-based method. The repetitive patterns and similarities are exploited to rectify geometrical and topological inaccuracies of the reconstructed models. Apart from developing the 3D façade model reconstruction scheme, the research focuses on individual tree segmentation and derivation of attributes of urban trees. The second method aims at extracting individual trees from the remaining point clouds. Knowledge about trees specially pertaining to urban areas is used in the process of tree extraction. An innovative shape based approach is developed to transfer this knowledge to machine language. The usage of principal direction for identifying stems is introduced, which consists of searching point segments representing a tree stem. The output of the algorithm is, segmented individual trees that can be used to derive accurate information about the size and locations of each individual tree. The reliability of the two methods is verified against three different data sets obtained from different laser scanner systems. The results of both methods are quantitatively evaluated using a set of measures pertaining to the quality of the façade reconstruction and tree extraction. The performance of the developed algorithms referring to the façade reconstruction, tree stem detection and the delineation of individual tree crowns as well as their limitations are discussed. The results show that MLS point clouds are suited to document urban objects rich in details. From the obtained results, accurate measurements of the most important attributes relevant to the both objects (building façades and trees), such as window height and width, area, stem diameter, tree height, and crown area are obtained acceptably. The entire approach is suitable for the reconstruction of building façades and for the extracting trees correctly from other various urban objects, especially pole-like objects. Therefore, both methods are feasible to cope with data of heterogeneous quality. In addition, they provide flexible frameworks, from which many extensions can be envisioned

This paper presents a study on classifying receipts and invoices using Machine Learning. Furthermore, Naïve Bayes Algorithm and the advantages of using it will be discussed.  With information gathered from theory and previous research, I will show how to classify images into a receipt or an invoice. Also, it includes pre-processing images using a variety of pre-processing methods and text extraction using Optical Character Recognition (OCR). Moreover, the necessity of pre-processing images to reach a higher accuracy will be discussed. A result shows a comparison between Tesseract OCR engine and FineReader OCR engine. After embracing much knowledge from theory and discussion, the results showed that combining FineReader OCR engine and Machine Learning is increasing the accuracy of the image classification.

Sensors based on LiDAR (Light Detection and Ranging) technology have the potential to provide accurate 3D models of the trees retrieving information such as canopy volume and height. This information can be used for diagnostics and prescriptions of fertilizers and plant protection products on a site-specific basis. This research aimed to investigate the use of LiDAR sensors in orange crops. Orange is one of the most important tree crop in Brazil. So far, research have developed and tested LiDAR based systems for several tree crops. However, usually individual trees or small field plots have been used. Therefore, several aspects related to data acquisition and processing must still be developed for large-scale application. The first study reported in this document (Chapter 3) aimed to develop and test a mobile terrestrial laser scanner (MTLS) and new data processing methods in order to obtain 3D models of large commercial orange groves and spatial information about canopy geometry. A 2D laser sensor and a RTK-GNSS receiver (Real Time Kinematics - Global Navigation Satellite System) were mounted on a vehicle. The data processing was based on generating a georeferenced point cloud, followed by the filtering, classification and surface reconstruction steps. A 25 ha commercial orange grove was used for field validation. The developed data acquisition and processing system was able to produce a reliable point cloud of the grove, providing high resolution canopy volume and height information. The choice of the type of point cloud classification (by individual trees or by transversal sections of the row) and the surface reconstruction algorithm is discussed in this study. The second study (Chapter 4) aimed to characterize the spatial variability of canopy geometry in commercial orange groves. Understanding such variability allows sensor-based variable rate application of inputs (i.e, applying proportional rates of inputs based on the variability of canopy size) to be considered as a suitable strategy to optimize the use of fertilizers and plant protection products. Five commercial orange groves were scanned with the developed MTLS system. According to the variability of canopy volume found in those groves, the input savings as a result of implementing sensor-based variable rate technologies were estimated in about 40%. The second goal of this study was to understand the relationship between canopy geometry and several other relevant attributes of the groves. The canopy volume and height maps of three groves were analyzed against historical yield maps, elevation, soil electrical conductivity, organic matter and clay content maps. The correlations found between canopy geometry and yield or soil maps varied from poor to strong correlations, depending on the grove. When classifying the groves into three classes according to canopy size, the yield performance and soil features inside each class was found to be significantly different, indicating that canopy geometry is a suitable variable to guide management zones delineation in one grove. Overall results from this research show the potential of MTLS systems and subsequent data analysis in orange crops indicating how canopy geometry information can be used in site-specific management practices.
Sensores baseados em tecnologia LiDAR (Light Detection and Ranging) têm o potencial de fornecer modelos tridimensionais de árvores, provendo informações como o volume e altura de copa. Essas informações podem ser utilizadas em diagnósticos e recomendações localizadas de fertilizantes e defensivos agrícolas. Este estudo teve como objetivo investigar o uso de sensores LiDAR na cultura da laranja, uma das principais culturas de porte arbóreo no Brasil. Diversas pesquisas têm desenvolvido sistemas LiDAR para culturas arbóreas. Porém, normalmente tais sistemas são empregados em plantas individuais ou em pequenas áreas. Dessa forma, diversos aspectos da aquisição e processamento de dados ainda devem ser desenvolvidos para viabilizar a aplicação em larga escala. O primeiro estudo deste documento (Capítulo 3) focou no desenvolvimento de um sistema LiDAR (Mobile Terrestrial Laser Scanner - MTLS) e nova metodologia de processamento de dados para obtenção de informações acerca da geometria das copas em pomares comerciais de laranja. Um sensor a laser e um receptor RTK-GNSS (Real Time Kinematics - Global Navigation Satellite System) foram instalados em um veículo para leituras em campo. O processamento de dados foi baseado na geração de uma nuvem de pontos, seguida dos passos de filtragem, classificação e reconstrução da superfície das copas. Um pomar comercial de laranja de 25 ha foi utilizado para a validação. O sistema de aquisição e processamento de dados foi capaz de produzir uma nuvem de pontos representativa do pomar, fornecendo informação sobre geometria das plantas em alta resolução. A escolha sobre o tipo de classificação da nuvem de pontos (em plantas individuais ou em seções transversais das fileiras) e sobre o algoritmo de reconstrução de superfície, foi discutida nesse estudo. O segundo estudo (Capítulo 4) buscou caracterizar a variabilidade espacial da geometria de copa em pomares comerciais. Entender tal variabilidade permite avaliar se a aplicação em taxas variáveis de insumos baseada em sensores LiDAR (aplicar quantias de insumos proporcionais ao tamanho das copas) é uma estratégia adequada para otimizar o uso de insumos. Cinco pomares comerciais foram avaliados com o sistema MTLS. De acordo com a variabilidade encontrada, a economia de insumos pelo uso da taxa variável foi estimada em aproximadamente 40%. O segundo objetivo desse estudo foi avaliar a relação entre a geometria de copa e diversos outros parâmetros dos pomares. Os mapas de volume e altura de copa foram comparados aos mapas de produtividade, elevação, condutividade elétrica do solo, matéria orgânica e textura do solo. As correlações entre geometria de copa e produtividade ou fatores de solo variaram de fraca até forte, dependendo do pomar. Quando os pomares foram divididos entre três classes com diferentes tamanhos de copas, o desempenho em produtividade e as características do solo foram distintas entre as três zonas, indicando que parâmetros de geometria de copa são variáveis úteis para a delimitação de unidades de gestão diferenciada em um pomar. Os resultados gerais desta pesquisa mostraram o potencial de sistemas MTLS para pomares de laranja, indicando como a geometria de copa pode ser utilizada na gestão localizada de pomares de laranja.

    The radio performance measurements of terminal antennas require expensive investments and they are often performed in anechoic chambers. Now a days measurement techniques using wheeler cap, reverberation chamber, etc have become popular. But these methods are used for measurements either during the design or production or in-network testing phases but not for all. Hence EMSCAN, a Canadian company, has come up with a device called Lab express near field scanner to counter this problem.

 

In this thesis an attempt is made to study the EMSCAN near field scanner by looking into architecture and design aspects of this device. Moreover, the capability of the device to measure the radio performance of the terminal antennas in terms of radiation pattern and total radiated power (TRP) is also studied and anlaysed.The analysis of the TRP and radiation pattern measurements of the terminal antennas is done by comparing the EMSCAN measurement results of 10 commercially available mobile phones with that of CTIA approved Satimo SG24 chamber.

 

The TRP and radiation pattern measurements are done for GSM 900 MHz and DCS 1800 MHZ low bands i.e.975 and 512 respectively. Furthermore, the dependency of the measured TRP on the positioning of the mobile phone on EMSCAN is tested by measuring the TRP and the radiated power of a single mobile phone.

 

The measurement results after comparison suggest that there is a correlation of around 79% at 1800 MHz and 49% at GSM 900 MHz among the two methods. This leads to the conclusion that EMSCAN scanner cannot replace the anechoic chambers for estimating the radio performance of terminal antennas. The results obtained by measuring the mobile phone in 8 different positions suggest that there is no significant difference in the TRP measured. There is a difference of 1.25dB was observed between the maximum and minimum TRP measured in two different positions on the near field scanner.

 

The results obtained by both the methods are verified by computing the TRP and radiation pattern in Matlab.The Matlab results agrees more with the EMSCAN results than Satimo SG 24 chamber results.

Painted artworks are complex multi-layered materials studied over the time with several sophisticated imaging techniques. Scanning MA-XRF technique provides information on pigment materials and their distribution over the painted surface in non-invasive manner. Further, it provides information on conservation state of artworks and also their authenticity can be approached. The penetrating nature of X-rays allows visualizing overpainted hidden pictorial compositions, not visible to naked eye. In the present PhD activity an advanced XRF scanning system for a real time imaging of painted artworks has been developed at the LANDIS laboratory of LNS-INFN and IBAM-CNR in Catania (Italy).The MA-XRF scanner presents a modular set-up to be assembled/disassembled in a short time (30-60 min) and easily moved in museums for analyses in situ. Spectrometric head is moved by three linear stages covering an area up to 110 × 70 × 20 cm3 (XYZ) . All motion sensors and their synchronization with the acquisition system are fully controlled by a custom developed and programmed Control Unit (CU) and all instrument operations can be in real-time monitored by a graphical user interface (GUI). The CU controls also the Z-axis movement, along the focal distance, via the communication with a laser triangulation sensor maintaining dynamically constant the painting-spectrometer distance in case of non-flat sample surface. Further, the CU interrupts immediately the scan if the safety distance is overcame. SDD detector coupled to a Digital X-ray Processors (DXP) that works in a time-list event mode (TLIST) allows to perform scans in real time (or in continuous mode). Ultra-fast scans can be performed up to 100 mm/sec covering the total area in 4.2 hours. The MA-XRF device is aimed to the macroscopic analysis of large size paintings. However, the use of a polycapillary lens equipping the X-ray source, allows to combine imaging at both the macro and micro scale of length with a simple adjustment of the sample position along the focal distance obtaining a lateral resolution up to 35 µm. A long range optical microscope is used for the accurate positioning of the painting. To date, performances of this device in terms of lateral resolution, scanning speed and dimensions of the scanning area, makes it as one of the most (or even the most) efficient portable MA-XRF systems available in the scientific community for studying paintings. X-ray spectra are processed in real-time during the scanning by the non-linear least-squares fitting procedure developed in PyMCa and integrated in the in-house programmed analysis software with a maximum processing speed of 5000 fitted spectra per second. Further a number of editing and mathematical tools for processing the images in live-mode are available. During the present research activity, the scanner was validated by studying three painted artworks on different support (canvas, wood and paper) and different size, for which the instrument has been moved in situ. MA-XRF provided valuable information through the determination of the pigment palette on painting techniques and artist connections with other workshops, the degradation state of the paintings, the reconstruction of the creative process by revealing overpainted figure and hidden inscriptions as well by the identification of changes introduced during the creation of the work. Analytical results obtained by MA-XRF analysis of the Paston Treasure , a mid-1670s oil on canvas painting at the Norwich Castle Museum (UK); painted wooden coffers belonging to the funerary collection of the tomb of Kha and Merit (1549-1292 BC) at the Museo Egizio in Torino (Italy); and an Italian illuminated manuscripts manufactured in the 14th century at the Fitzwilliam Museums (Cambridge) are presented and discussed.

La cartographie mobile est un sujet qui se démocratise et gagne en maturité avec la demande croissante des données tridimensionnelles urbaines et périurbaines. La présente thèse aborde la conception d’un système de cartographie mobile terrestre à faible coût avec la particularité d’avoir adapté un scanner laser terrestre pour une utilisation en mode mobile à faible vitesse. Notre objectif n’est pas de rivaliser en termes de performances avec les systèmes commerciaux mais plutôt de s’approprier les compétences scientifiques et technologiques qui permettront de proposer des solutions dans le domaine de cartographie mobile. Les opérations préalables à la mise en route du système, tel que la synchronisation et l’étalonnage sont exposées. Puis, des méthodes basées sur l’ajustement des modèles polynomiaux sont développées pour convenir aux différents trajets parcourus. Les données issues des différents capteurs (GPS/AHRS/TLS) sont testées et filtrées avant de les intégrer dans l’équation de géoréférencement direct. Il en résulte un nuage de points corrigé. Une étude exhaustive sur l’effet des erreurs de chaque capteur sur le nuage de points résultant est établie. La précision théorique est confrontée avec des jeux des données de référence pour valider l’analyse des erreurs. Un appareil photo numérique calibré est intégré dans notre système en tant que capteur de navigation. Une solution photogrammétrique est proposée pour améliorer la précision de la trajectométrie préalablement calculée par l’intégration des données GPS/AHRS. Pour conclure la thèse, une ouverture vers la modélisation géométrique est proposée pour exploiter la géométrie et la précision des données laser terrestres fournies par le système. Le prototype développé permet de fournir des nuages de points d’une précision de l’ordre de 10 à 15cm à une distance moyenne de 20 m
Mobile mapping technology has been developing with the growing demand of three-dimensional urban and peri-urban data. This thesis approach is based on the design of a low cost terrestrial mobile mapping system with the adaptation of a Terrestrial Laser Scanner for low dynamics. Our goal is not to compete in performance with commercial systems but rather to appropriate scientific and technological skills which will help in proposing solutions in the field of mobile mapping. Necessary operational settings, such as synchronization and calibration are explained. Then, some methods based on the adjustment of polynomial models are developed according to the traveled paths. Data from various sensors (GPS/ AHRS/TLS) are filtered and tested before their integration by direct georeferencing equation in order to produce a correct point cloud. A comprehensive study on the influence of errors of each sensor on the resulting point cloud is established. The theoretical precision is compared with reference data in order to validate the error analyze. A digital calibrated camera is integrated in the system as a navigation sensor. A photogrammetric solution is proposed to improve the accuracy of the orientation and the position calculated by integrating GPS/ AHRS. At the end of this thesis, an approach towards automatic modeling is proposed to make use of the geometry and precision provided by the system. The designed prototype supplies point clouds whose precision is about 10 to15 cm at the average distance of 20 m

Mestrado em Engenharia Florestal e dos Recursos Naturais / Instituto Superior de Agronomia. Universidade de Lisboa
LiDAR (Light Detection And Ranging) é um sistema baseado nos princípios de Deteção Remota que permite medir distâncias com base no tempo da trajetória da radiação laser, desde que é emitida pelo aparelho até que retorna ao recetor depois de ser refletida numa superfície sólida. A aplicabilidade deste sistema é abrangente a várias áreas da engenharia e prende-se com a capacidade que o mesmo tem de recolher e armazenar dados tridimensionais em forma de nuvens de pontos de qualquer objeto sólido sobre a superfície terrestre. No sector florestal, este sistema permite estimar características dos povoamentos e digitalizar uma extensa área de floresta, de uma forma automatizada, rápida e com detalhe na ordem dos milímetros. O objetivo do presente trabalho é avaliar a capacidade do sistema LiDAR na individualização da árvore comparando as coordenadas estimadas obtidas com dois métodos LiDAR - HMLS (Held-Hand Mobile Laser Scanner) e TLS (Terrestrial Taser Scanner) - com as coordenadas obtidas com GPS sub-métrico, pelo método tradicional de campo, num ensaio clonal de Eucalyptus globulus Labill. com 10 anos de idade. O presente estudo serviu também como primeira abordagem ao desempenho dos dois métodos LiDAR na obtenção de diâmetros às várias alturas do tronco, recorrendo aos algoritmos disponíveis no software R. Para a deteção das árvores individuais, os resultados demonstraram que, em média, o método TLS detetou 65,1% das árvores, enquanto o método HMLS detetou 44,7% das árvores, para todas as parcelas de estudo. Comprovou-se ainda que o levantamento com HMLS só é vantajoso para terrenos regulares e percursos retos. Concluiu-se que deve ser efetuada uma melhoria nos processos associados à utilização do algoritmo SLAM (Simultaneous Localization And Mapping) e salientou-se a importância de utilizar pontos de referência em campo para a obtenção de nuvens de pontos de melhor qualidade
N/A

Heritage Building Information Modelling is strongly connected with the need for accurate surveys, both because BIM models are not usually available and because of the historical buildings’ peculiarity. But acquiring those data demands the involvement of additional costly stakeholders, and that limits the widespread use of LiDAR technologies. New smart devices supplied with this technology may allow customers to acquire point clouds with a different approach, giving us more and more possibilities of easily acquiring a point cloud, possibly omitting the use of expensive equipment and the stakeholders (the surveyors) involved in their use; for example, the latest hardware installed by Apple on their devices come equipped with a LiDAR sensor and their patented True Depth algorithm, offering cheap alternatives in acquiring 3D scans. This thesis wants to deal with the comparison of different approaches of acquiring a point cloud, by exploring multiples possibilities and risks of using this new technology, starting from a review of the scientific literature on the subject, and then comparing the results of the different methodologies by using different point clouds realized by devices in three price range (namely a BLK2GO, a LEICA P20 laser scanner and an iPhone 12 PRO) using a series of buildings as case studies, focusing on heritage buildings. For this purpose, a series of scans will be analysed metrically and quantitatively, to understand their accuracy and recognize which level of accuracy (LOA) can be reached, and so understand which purpose this new low-priced technologies can be used for.

Ce travail de thèse porte sur la numérisation complète d'environnements vastes et complexes en utilisant un TLS ou un scanner mobile. Nous proposons un planificateur de vue multi-objectif adaptatif capable de fonctionner dans un environnement inconnu pour guider l'opérateur humain en mode "hors ligne" et faciliter la tâche de numérisation ou en mode "en ligne" avec un scanner embarqué sur un robot mobile pour une exploration automatique de l'environnement. La méthode proposée suppose que le scanner est déplacé sur une surface plane ce qui est fréquent dans les environnements d'intérieur, les zones urbaines, les espaces ouverts ou encore dans certaines applications du patrimoine culturel. Dans un premier temps, nous proposons une nouvelle stratégie d'exploration automatisée qui exploite des régions spécifiques de l'environnement, nommées "cellules conservatrices" pour réduire le nombre de déplacements du scanner et obtenir une numérisation complète de l'environnement. Ensuite, nous introduisons une approche pour améliorer le processus de numérisation en mode "hors ligne", en particulier lors de l'utilisation d'un TLS dans les grands environnements. Pour ce faire, nous avons proposé de combiner l'utilisation d'un scanner laser terrestre avec un robot mobile équipé d'un miroir plan. La méthode permet de réduire considérablement l'effort fourni par l'opérateur humain pour déplacer le scanner dans l'environnement et améliore le taux de complétude du nuage de points final. Les méthodes proposées dans cette thèse ont été validées avec des nuages de points simulés et réels avec un TLS et un robot mobile et montrent de bonnes performances en termes de taux de couverture et de temps de calcul par rapport à d'autres stratégies de planification de vues dans la littérature, mais aussi par rapport aux résultats obtenus par un opérateur humain expérimenté dans un environnement large et complexe
This PhD work investigates automatic digitization with complete coverage of large and complex environments using a terrestrial laser scanner (TLS) or a mobile scanner. We propose an adaptive multi-objective view-planner that can operate in an unknown environment to provide in offline mode guidance for the human operator and ease the scanning task or in online mode with a scanner embedded on a mobile robot for an automatic exploration of the environment. The proposed method assumes that the laser scanner is moved on a flat surface which is common in indoor environments, urban areas, open spaces or in various cultural heritage applications. First, we propose a novel exploration strategy that is completely automated and does not require extensive computations that uses specific regions of the environment called "Conservative-Cells" to drastically reduce the number of sensing positions to achieve complete digitization of the environment. Next, we present an approach to improve the scanning process in "offline" mode, especially when using a TLS in large environments. For this purpose, we suggested combining the use of a terrestrial laser scanner with a mobile robot equipped with a planar mirror. The result is a significant reduction in the effort required by the human operator to move the scanner in the environment and improve the completeness rate of the final point cloud. Proposed methods were validated with simulated and real point clouds on both TLS and mobile robot. The proposed approaches show efficient performance in terms of coverage rate and computational time compared to other view-planning approaches as well as the results of an experienced human operator in a large, complex environment

Cette thèse s’inscrit dans le cadre d’un projet industriel multi-partenaires ayant pour objectif le développement d’un robot mobile collaboratif (un cobot), autonome dans ses mouvements au sol, capable de réaliser l’inspection visuelle d’un aéronef, à la fois en phase de petite ou grande maintenance dans un hangar ou en phase de pré-vol sur le tarmac d’un aéroport. Le cobot est équipé de capteurs lui permettant d’effectuer ses tâches de navigation autonome, mais également d’un ensemble de capteurs optiques constituant la tête d’inspection : une caméra orientable Pan-Tilt-Zoom et un scanner 3D qui délivrent respectivement des données sous forme d’images 2D et de nuages de points 3D. L’objectif de la thèse est de développer des algorithmes d’analyse d’images 2D et de nuages de points 3D, afin d’établir un diagnostic sur l’état de l’avion et son aptitude à voler. Nous avons développé des algorithmes pour vérifier certains éléments de l’appareil, tels que valves, portes, capteurs, pneus ou moteurs, et également pour détecter et caractériser des dommages 3D sur le fuselage (impacts, rayures, etc.). Nous avons exploité dans nos algorithmes les connaissances a priori disponibles, en particulier le modèle 3D CAO de l’avion (un AIRBUS A320 dans le cadre de nos essais). Durant ces travaux de la thèse, nous avons pu répondre à deux besoins (parfois antagonistes) : développer des algorithmes d’inspection rapides et robustes, mais également répondre aux exigences spécifiques d’un projet industriel qui visait à développer un prototype opérationnel. Nous nous sommes attachés à développer des algorithmes les plus génériques possibles, de manière à ce qu’ils puissent être utilisés pour d’autres types d’inspection, tels que l’inspection de bâtiments ou de navires par exemple. Nous avons aussi contribué au développement du prototype (robot mobile équipé de capteurs) en développant le module de contrôle des capteurs d’inspection et en intégrant nos codes sur le robot avec les autres modules développés par les partenaires. Le prototype a fait l’objet de nombreux essais en hangar de maintenance ou sur tarmac
This thesis makes part of an industry oriented multi-partners project aimed at developing a mobile collaborative robot (a cobot), autonomous in its movements on the ground, capable of performing visual inspection of an aircraft during short or long maintenance procedures in the hangar or in the pre-flight phase on the tarmac. The cobot is equipped with sensors for realizing its navigation tasks as well as with a set of optical sensors which constitute the inspection head: an orientable Pan-Tilt-Zoom visible light camera and a three-dimensional scanner, delivering data in the format of two-dimensional images and three-dimensional point clouds, respectively. The goal of the thesis is to propose original approaches for processing 2D images and 3D clouds, with intention to make a decision with respect to the flight readiness of the airplane. We developed algorithms for verification of the aircraft items such as vents, doors, sensors, tires or engine as well as for detection and characterization of three-dimensional damages on the fuselage. We integrated a-priori knowledge on the airplane structure, notably numerical three-dimensional CAD model of the Airbus-A320. We argue that with investing effort to develop robust enough algorithms and with the help of existing optical sensors to acquire suitable data, we can come up with non-invasive, accurate, and time-efficient system for automatic airplane exterior inspection. The thesis work was placed in between two main requirements: develop inspection algorithms which could be as general as possible and also meet the specific requirements of an industry oriented project. Often, these two goals do not go along and the balance had to be made. On one side, we were aiming to design and assess the approaches that can be employed on other large structures, for ex. buildings, ships. On the other hand, writing source code for controlling sensors as well as integrating our whole developed source code with other modules on the real-time robotic system, were necessary in order to demonstrate the feasibility of our robotic prototype

Cette thèse a été effectuée en collaboration entre le Centre de Robotique CAOR de MINES ParisTech et MENSI-Trimble sous convention CIFRE (Convention Industrielle de Formation par la Recherche), afin de concevoir un nouveau système de scanner laser innovant dédié aux systèmes mobiles de cartographie. Pour cela, nous avons développé une méthode d'analyse qualitative des relevés laser et une démarche de conception par simulation d'un système mobile de cartographie. Nous avons ensuite élaboré plusieurs concepts de scanners laser mobiles. A l'aide du simulateur, nous avons virtuellement réalisé ces concepts et avons fait des acquisitions simulées afin d'analyser les données. Après avoir obtenu les résultats de l'analyse, nous avons choisi un concept, réalisé son prototypage et effectué son évaluation en situation réelle
This thesis is a collaboration between the Robotics Centre CAOR of MINES ParisTech and MENSI-Trimble under the CIFRE (Convention Industrielle de Formation par la Recherche) convention, to desgin an innovative laser scanning system dedicated to mobile mapping systems. We developed a method for the qualitative analysis of point-cloud data and the conception of a mobile mapping system using simulation. We elaborated several concepts for mobile laser scanners, which we realized virtually using the simulator. We performed data acquisitions in order to analyze the simulated data. We chose the most suitable concept, based on our analysis results, constructed a prototype and assessed its performance in a real environment

Cette thèse a été effectuée en collaboration entre le Centre de Robotique CAOR de MINES ParisTech et MENSI-Trimble sous convention CIFRE (Convention Industrielle de Formation par la Recherche), afin de concevoir un nouveau système de scanner laser innovant dédié aux systèmes mobiles de cartographie. Pour cela, nous avons développé une méthode d'analyse qualitative des relevés laser et une démarche de conception par simulation d'un système mobile de cartographie. Nous avons ensuite élaboré plusieurs concepts de scanners laser mobiles. A l'aide du simulateur, nous avons virtuellement réalisé ces concepts et avons fait des acquisitions simulées afin d'analyser les données. Après avoir obtenu les résultats de l'analyse, nous avons choisi un concept, réalisé son prototypage et effectué son évaluation en situation réelle.

This master’s thesis describes a sensorial subsystem of a mobile robot. The thesis mentions the structure and a basis of the satellite navigation. There are also described each systems and visualization techniques as e. g. WGS 84, including also other systems for specification of more accurate position on the Earth (EGNOS, WAAS and others). This thesis describes closely also a differential GPS and the corrections. For the purposes of the thesis there has been assembled GPS software for a fundamental and a remote station. There have been chosen and used modules for receiving a GPS signal and wireless modules for their communication. There are also used outcomes from the project of the Open Street Map in the thesis. Results of this project are described in details, decoded, adjusted and converted into RNDF file. At the end of this thesis there has been composed a function of the program for portable camera which works on principle time of flight. This portable camera is dedicated to look for immediate barrier and also for 3D space view in front of the robot. This thesis also including the process how the power source has been built up for this portable scanner.

Die Kartierung und Lokalisierung eines mobilen Roboters in seiner Umgebung ist eine wichtige Voraussetzung für dessen Autonomie. In dieser Arbeit wird der Einsatz eines 3D-Laserscanners zur Erfüllung dieser Aufgaben untersucht. Durch die optimierte Anordnung eines rotierenden 2D-Laserscanners werden hochauflösende Bereiche vorgegeben. Zudem wird mit Hilfe von ICP die Kartierung und Lokalisierung im Stillstand durchgeführt. Bei der Betrachtung zur Verbesserung der Bewegungsschätzung wird auch eine Möglichkeit zur Lokalisierung während der Bewegung mit 3D-Scans vorgestellt. Die vorgestellten Algorithmen werden durch Experimente mit realer Hardware evaluiert.

Thesis deals with data fusion and calibration of sensory system of Orpheus-X3 robot and EnvMap mapping robot. These robots are parts of Cassandra robotic system that is used for exploration of hazardous or inaccessible areas. Corrections of measured distances are determined for used laser scanners Velodyne HDL-64, Velodyne HDL-32 and range camera SwissRanger SR4000. Software MultiSensCalib has been created and is described. This software is used for determination of intrinsic parameters of heterogeneous cameras of the sensory head and for determination of mutual position and orientation of these sensors. Algorithm for data fusion of CCD camera stereo pair, thermal imager stereo pair and range camera is proposed. Achieved calibration and data-fusion parameters are evaluated by several experiments.

Im Rahmen einer Diplomarbeit an der TU Bergakademie Freiberg wurden in 2014 die Grundlagen für die Auswertung von 3D-Punktwolken zur automatisierten Nachtragung des Risswerks gelegt. Um die dafür notwendigen 3D-Punktwolken möglichst wirtschaftlich zu erstellen, laufen seit 2015 Untersuchungen und Testmessungen zur Machbarkeit des untertägigen Einsatzes von mobil messenden Laserscannern. Im Folgenden werden verschiedene technische Ansätze sowie die Ergebnisse der Testmessungen und die weiteren geplanten Schritte vorgestellt
As part of a thesis at the Technical University of Freiberg, a basis for the analysis of 3D point clouds was set for refining the mine map automatically. Since 2015 studies and test measurements have been running to create the necessary 3D point clouds as economically as possible, by using an underground mobile scanning system. Below the different technical approaches will be presented as well as the results of the test measurements and the next planned steps

All’interno del patrimonio nazionale dei Beni Culturali, si inserisce la categoria delle strutture ipogee, numerosissime nel territorio italiano: ne sono un esempio i percorsi ipogei presenti a Santarcangelo di Romagna, oggetto della presente tesi. Le moderne tecniche geomatiche di rilievo tridimensionale, specialmente se integrate tra loro, offrono valide e molteplici soluzioni per la documentazione dello stato di fatto di tali ambienti sotterranei. La sperimentazione condotta nell’ambito della presente tesi di laurea ha previsto l’esecuzione di una rete topografica di precisione e l’impiego della tecnica laser a scansione terrestre, per il rilievo 3D e la georeferenziazione di una porzione del complesso ipogeo della città di Santarcangelo di Romagna. I dati di rilievo acquisiti (misure topografiche e modello 3D a nuvola di punti) e gli elaborati grafici a valenza metrica ottenuti sono funzionali ad eventuali attività di monitoraggio e restauro delle cavità ipogee. Nella conclusione della tesi vengono prospettate possibili soluzioni per estendere in futuro l’implementazione del rilievo all’intero complesso ipogeo, adottando le più moderne tecnologie di mobile mapping.

La localisation automatique est une fonctionnalité importante des systèmes de cartographie mobiles (Mobile Mapping Systems, MMS). La présente thèse présente des solutions complémentaires aux méthodes de localisation utilisées actuellement dans un système MMS terrestre, qui utilise des récepteurs GPS et des centrales à inertie (Inertial Measurement Units, IMU). Un post-traitement, par lissage des données, permet d'améliorer les cartes 3D générées par un MMS. Cette approche est cependant insuffisante pour corriger les erreurs à variations lentes des capteurs. La présente thèse propose une technique de localisation alternative, fondée sur des scanners 2D à lasers. La méthode présentée ici, d'odométrie par laser, utilise des repères plans, qui sont fréquents dans les environnements créés par l'Homme : ces repères fixes permettent de déterminer le déplacement opéré par la plateforme mobile. Contrairement à la technique du SLAM (Simultaneous Localization and Mapping), utilisée pour la navigation des robots à l'intérieur d'un bâtiment, la transformation 3D est calculée sans avoir recours à une carte préétablie, mais en exploitant des propriétés invariantes des caractéristiques extraites de l'environnement. Nous proposons une approche par "division pour régner" (divide and conquer, D&C) qui simplifie les tâches d'association des repères (data association, DA) et de reconstruction du mouvement.

Aktuelle 3D-Stadtmodelle werden immer wichtiger in verschiedenen städtischen Anwendungsbereichen. Im Moment dienen sie als Grundlage bei der Stadtplanung, virtuellem Tourismus und Navigationssystemen. Mittlerweile ist der Bedarf an 3D-Gebäudemodellen dramatisch gestiegen. Der Grund dafür sind hauptsächlich Navigationssysteme und Onlinedienste wie Google Earth. Die Mehrheit der Untersuchungen zur Rekonstruktion von Gebäudemodellen von Luftaufnahmen konzentriert sich ausschließlich auf Dachmodellierung. Jedoch treiben Anwendungen wie Virtuelle Realität und Navigationssysteme die Nachfrage nach detaillieren Gebäudemodellen, die nicht nur die geometrischen Aspekte sondern auch semantische Informationen beinhalten, stark an. Urbanisierung und Industrialisierung beeinflussen das Wachstum von urbaner Vegetation drastisch, welche als ein wesentlicher Teil des Lebensraums angesehen wird. Aus diesem Grund werden Aufgaben wie der Ökosystemüberwachung, der Verbesserung der Planung und des Managements von urbanen Regionen immer mehr Aufmerksamkeit geschenkt. Gleichermaßen hat die Erkennung und Modellierung von Bäumen im Stadtgebiet sowie die kontinuierliche Überprüfung ihrer Inventurparameter an Bedeutung gewonnen. Die steigende Nachfrage nach 3D-Gebäudemodellen, welche durch Fassadeninformation ergänzt wurden, und Informationen über einzelne Bäume im städtischen Raum erfordern effiziente Extraktions- und Rekonstruktionstechniken, die hochgradig automatisiert sind. In diesem Zusammenhang ist das Wissen über die geometrische Form jedes Objektteils ein wichtiger Aspekt. Heutzutage, wird das Mobile Laser Scanning (MLS) vermehrt eingesetzt um Objekte im städtischen Umfeld zu erfassen und es entwickelt sich zur Hauptquelle von Daten für die Modellierung von urbanen Objekten. Eine Vielzahl von Objekten wurde schon mit Daten von MLS rekonstruiert. Außerdem wurden bereits viele Methoden für die Verarbeitung von MLS-Daten mit dem Ziel urbane Objekte zu erkennen und zu rekonstruieren vorgeschlagen. Die 3D-Punkwolke einer städtischen Szene stellt eine große Menge von Messungen dar, die viele Objekte von verschiedener Größe umfasst, komplexe und unvollständige Strukturen sowie Löcher (Rauschen und Datenlücken) enthält und eine inhomogene Punktverteilung aufweist. Aus diesem Grund ist die Verarbeitung von MLS-Punktwolken im Hinblick auf die Extrahierung und Modellierung von wesentlichen und charakteristischen Fassadenstrukturen sowie Bäumen von großer Bedeutung. In der Arbeit werden zwei neue Methoden für die Rekonstruktion von Gebäudefassaden und die Extraktion von Bäumen aus MLS-Punktwolken vorgestellt, sowie ihre Anwendbarkeit in der städtischen Umgebung analysiert. Die erste Methode zielt auf die Rekonstruktion von Gebäudefassaden mit expliziter semantischer Information, wie beispielsweise Fenster, Türen, und Balkone. Die Rekonstruktion läuft vollautomatisch ab. Zu diesem Zweck werden einige Algorithmen vorgestellt, die auf dem Vorwissen über die geometrische Form und das Arrangement von Fassadenmerkmalen beruhen. Die initiale Klassifikation, mit welcher die Punkte in Objektpunkte und Bodenpunkte unterschieden werden, wird über eine lokale Höhenhistogrammanalyse zusammen mit einer planaren Region-Growing-Methode erzielt. Die Punkte, die als zugehörig zu Objekten klassifiziert werden, werden anschließend in Ebenen segmentiert, welche als Basiselemente der Merkmalserkennung angesehen werden können. Information über die Gebäudestruktur kann in Form von Regeln und Bedingungen erfasst werden, welche die wesentlichen Steuerelemente bei der Erkennung der Fassadenmerkmale und der Rekonstruktion des geometrischen Modells darstellen. Um Merkmale wie Fenster oder Türen zu erkennen, die sich an der Gebäudewand befinden, wurde eine löcherbasierte Methode implementiert. Einige Löcher, die durch Verdeckungen entstanden sind, können anschließend durch einen neuen regelbasierten Algorithmus eliminiert werden. Außenlinien der Merkmalsränder werden durch ein Polygon verbunden, welches das geometrische Modell repräsentiert, indem eine Methode angewendet wird, die auf geometrischen Primitiven basiert. Dabei werden die topologischen Relationen unter Beachtung des Vorwissens über die primitiven Formen analysiert. Mögliche Außenlinien können von den Kantenpunkten bestimmt werden, welche mit einer winkelbasierten Methode detektiert werden können. Wiederkehrende Muster und Ähnlichkeiten werden ausgenutzt um geometrische und topologische Ungenauigkeiten des rekonstruierten Modells zu korrigieren. Neben der Entwicklung des Schemas zur Rekonstruktion des 3D-Fassadenmodells, sind die Segmentierung einzelner Bäume und die Ableitung von Attributen der städtischen Bäume im Fokus der Untersuchung. Die zweite Methode zielt auf die Extraktion von individuellen Bäumen aus den Restpunktwolken. Vorwissen über Bäume, welches speziell auf urbane Regionen zugeschnitten ist, wird im Extraktionsprozess verwendet. Der formbasierte Ansatz zur Extraktion von Einzelbäumen besteht aus einer Reihe von Schritten. In jedem Schritt werden Objekte in Abhängigkeit ihrer geometrischen Merkmale gefunden. Stämme werden unter Ausnutzung der Hauptrichtung der Punktverteilung identifiziert. Dafür werden Punktsegmente gesucht, die einen Teil des Baumstamms repräsentieren. Das Ergebnis des Algorithmus sind segmentierte Bäume, welche genutzt werden können um genaue Informationen über die Größe und Position jedes einzelnen Baumes abzuleiten. Einige Beispiele der Ergebnisse werden in der Arbeit angeführt. Die Zuverlässigkeit der Algorithmen und der Methoden im Allgemeinen wurden unter Verwendung von drei Datensätzen, die mit verschiedenen Laserscannersystemen aufgenommen wurden, verifiziert. Die Untersuchung zeigt auch das Potential sowie die Einschränkungen der entwickelten Methoden wenn sie auf verschiedenen Datensätzen angewendet werden. Die Ergebnisse beider Methoden wurden quantitativ bewertet unter Verwendung einer Menge von Maßen, die die Qualität der Fassadenrekonstruktion und Baumextraktion betreffen wie Vollständigkeit und Genauigkeit. Die Genauigkeit der Fassadenrekonstruktion, der Baumstammdetektion, der Erfassung von Baumkronen, sowie ihre Einschränkungen werden diskutiert. Die Ergebnisse zeigen, dass MLS-Punktwolken geeignet sind um städtische Objekte detailreich zu dokumentieren und dass mit automatischen Rekonstruktionsmethoden genaue Messungen der wichtigsten Attribute der Objekte, wie Fensterhöhe und -breite, Flächen, Stammdurchmesser, Baumhöhe und Kronenfläche, erzielt werden können. Der gesamte Ansatz ist geeignet für die Rekonstruktion von Gebäudefassaden und für die korrekte Extraktion von Bäumen sowie ihre Unterscheidung zu anderen urbanen Objekten wie zum Beispiel Straßenschilder oder Leitpfosten. Aus diesem Grund sind die beiden Methoden angemessen um Daten von heterogener Qualität zu verarbeiten. Des Weiteren bieten sie flexible Frameworks für das viele Erweiterungen vorstellbar sind.
Up-to-date 3D urban models are becoming increasingly important in various urban application areas, such as urban planning, virtual tourism, and navigation systems. Many of these applications often demand the modelling of 3D buildings, enriched with façade information, and also single trees among other urban objects. Nowadays, Mobile Laser Scanning (MLS) technique is being progressively used to capture objects in urban settings, thus becoming a leading data source for the modelling of these two urban objects. The 3D point clouds of urban scenes consist of large amounts of data representing numerous objects with significant size variability, complex and incomplete structures, and holes (noise and data gaps) or variable point densities. For this reason, novel strategies on processing of mobile laser scanning point clouds, in terms of the extraction and modelling of salient façade structures and trees, are of vital importance. The present study proposes two new methods for the reconstruction of building façades and the extraction of trees from MLS point clouds. The first method aims at the reconstruction of building façades with explicit semantic information such as windows, doors and balconies. It runs automatically during all processing steps. For this purpose, several algorithms are introduced based on the general knowledge on the geometric shape and structural arrangement of façade features. The initial classification has been performed using a local height histogram analysis together with a planar growing method, which allows for classifying points as object and ground points. The point cloud that has been labelled as object points is segmented into planar surfaces that could be regarded as the main entity in the feature recognition process. Knowledge of the building structure is used to define rules and constraints, which provide essential guidance for recognizing façade features and reconstructing their geometric models. In order to recognise features on a wall such as windows and doors, a hole-based method is implemented. Some holes that resulted from occlusion could subsequently be eliminated by means of a new rule-based algorithm. Boundary segments of a feature are connected into a polygon representing the geometric model by introducing a primitive shape based method, in which topological relations are analysed taking into account the prior knowledge about the primitive shapes. Possible outlines are determined from the edge points detected from the angle-based method. The repetitive patterns and similarities are exploited to rectify geometrical and topological inaccuracies of the reconstructed models. Apart from developing the 3D façade model reconstruction scheme, the research focuses on individual tree segmentation and derivation of attributes of urban trees. The second method aims at extracting individual trees from the remaining point clouds. Knowledge about trees specially pertaining to urban areas is used in the process of tree extraction. An innovative shape based approach is developed to transfer this knowledge to machine language. The usage of principal direction for identifying stems is introduced, which consists of searching point segments representing a tree stem. The output of the algorithm is, segmented individual trees that can be used to derive accurate information about the size and locations of each individual tree. The reliability of the two methods is verified against three different data sets obtained from different laser scanner systems. The results of both methods are quantitatively evaluated using a set of measures pertaining to the quality of the façade reconstruction and tree extraction. The performance of the developed algorithms referring to the façade reconstruction, tree stem detection and the delineation of individual tree crowns as well as their limitations are discussed. The results show that MLS point clouds are suited to document urban objects rich in details. From the obtained results, accurate measurements of the most important attributes relevant to the both objects (building façades and trees), such as window height and width, area, stem diameter, tree height, and crown area are obtained acceptably. The entire approach is suitable for the reconstruction of building façades and for the extracting trees correctly from other various urban objects, especially pole-like objects. Therefore, both methods are feasible to cope with data of heterogeneous quality. In addition, they provide flexible frameworks, from which many extensions can be envisioned.

碩士
朝陽科技大學
資訊工程系
104
3D image reconstruction has important applications in many fields such as Virtual Reality, E-commerce, 3D Game, and etc. The traditional 3D scanner device has 3D reconstruction of the effect of high accuracy degree, but it’s equipment is expensive. This paper proposes a three-dimensional reconstruction of smart devices based on a method for low cost and convenience features. Basic configuration is a smart device with a rotating disc. The use of smart devices on both the sensor and the direction of the camera function automatically measure the direction the camera angle information and equipment between the relationship, providing great convenience in image reconstruction, this method can be used to any smart device has a direction sensor and camera. In the basic configuration, only two three-dimensional reconstruction calculation methods are available. Contour method which the edge of the object can not be rendered in contour line, resulting in lost details. The stereo vision method susceptible to repeated low texture and texture, resulting in three-dimensional reconstruction results with the real object far. In this study, under the concept of a virtual point cloud space, using the projector and the back projection principle to combine the advantages of a successful method and stereoscopic contour method also overcomes the disadvantages of both. Experimental results show the proposed method overcomes the contour method and binocular vision in the overall error. In the future other than basic configuration can also be coupled with a light source, laser, and other expansion devices, so that it can join the other three-dimensional reconstruction method to enhance the effects of reconstruction.

In image-guided surgical navigation, instruments tools are tracked by a position sensor and their locations rendered along with the patient's anatomy. Conventional methods require an invasive, time-consuming and potentially uncertain process of intra-operative registration of the images to the patient. In a direct navigation system, such an intra-operative registration is replaced with pre-operative patient-independent calibration in a process that determines the relationship between the coordinate frame of imaging equipment and the coordinate frame of the position sensor. This dissertation presents a method for pre-operatively calibrating a direct navigation system that used an optical position sensor and a mobile gantry CT scanner. A custom bi-local calibration device was designed and manufactured, after which existing navigation software was augmented with components that used the pre-operatively determined transformation to provide image-guided surgical navigation. The resulting system was tested in an image-guided operating suite using plastic bone models. In the validation stage, the inherent error was less than 0.4 mm and the target registration error was approximately 1.6 mm for a ceiling-mounted position sensor and 0.7 mm for a portable position sensor. This accuracy is consistent with the best intra-operative registrations reported in the literature and this calibration method may be useful in future surgical systems.
Thesis (Master, Computing) -- Queen's University, 2013-09-05 12:53:08.994

碩士
國立臺灣科技大學
電機工程系
96
The purpose of this study is to propose the wheeled mobile robot (WMR) navigation system under an unknown environment, which consisting the 2D distance detection function, the environmental outline and the route constructing functions. The environmental outlines, which are the distance information in the front of the robot within the segmental area of the 180 degrees, are constructed based on the collecting angles and distance values through the detection and examination of the laser range finder. The robot also can move towards the specified goal progressively, and draw the outline of the surrounding after applying the fuzzy logic based regional path planning function. The environmental route mapping database from the starting point to the goal point can be created step by step with the detecting and analyzing the received individual map. In addition, this study adopts the mapping template to compare with the collected environmental information and a compass is also equipped to compensate the robot slipping to guide the robot’s movement more precisely.

碩士
國立中興大學
電機工程學系
92
This thesis develops methodologies and techniques for pose-tracking and pose initialization of an autonomous mobile robot (AMR) using two different external sensors: ultrasonics and laser scanner. First, a novel FPGA-based ultrasonic pose-tracking system by fusing the time-of-flight (TOF) readings together with the FAEIF algorithm is proposed to improve the accuracy and robustness of pose estimation for the AMR. The FAEIF-based sensor fusion approach is presented to circumvent the nonlinear filter divergence problems. Second, a low-complexity and accurate pose-tracking EKF algorithm is proposed based on rectangular model and a 2-D laser scanner. Finally, a pose initialization scheme based on a 2-D laser scanner is presented to determine the initial pose of the AMR given that the environmental model is known. Numerous simulations and experimental results are provided to verify the feasibility and effectiveness of the proposed pose-tracking and pose initialization methods.

A 20 anni dalla nomina di Patrimonio UNESCO della città veronese, il presente lavoro si concentra sulla salvaguardia del Patrimonio della città scaligera attraverso le più recenti tecnologie digitali sviluppatesi negli ultimi decenni. La via della conservazione non può che partire da quella della conoscenza e il rilievo delle mura di Verona, processo inedito sul territorio scaligero, è stato così occasione di verifica di un procedimento di rilievo integrato dove gli sviluppi tecnologici messi a disposizione nel campo della documentazione, danno notevoli opportunità per il rilevamento dell’Architettura e soprattutto per il rilievo nel settore dei Beni Culturali, sia per quanto riguarda l’acquisizione delle informazioni riferite all’oggetto d’indagine, cioè la fase di acquisizione del dato metrico, sia per quanto riguarda la questione della rappresentazione e divulgazione. Il progetto di ricerca Verona Fortificata, riguardo alla documentazione del sito UNESCO di Verona, nasce nel 2015 grazie alla collaborazione dell’Università degli Studi di Firenze e dell’Università degli Studi di Pavia, con il Comune di Verona e la partecipazione di professori, studiosi e ricercatori del settore. Nell’ottica di costruire, per la prima volta sul territorio veronese, una documentazione digitale completa del Patrimonio UNESCO, durante le varie campagne di rilievo svolte tra il 2015 e il 2019, sono stati acquisiti un grande mole di contenuti riguardanti parte della cinta muraria e delle fortificazioni presenti sul territorio scaligero. La raccolta ordinata degli stessi va ad assumere un punto chiave per la riuscita del progetto, per la sua facile comprensione e condivisione dei suoi risultati. La ricerca si divide in tre parti, di cui la prima ha carattere introduttivo, conoscitivo proponendosi di fornire le basilari informazioni storico-culturali imprescindibili al corretto inquadramento delle attività di ricerca e studio del perimetro scaligero; viene presentata una panoramica generale sulle vicende storiche che hanno portato la trasformazione nei secoli della città di Verona e le sue mura, che, tramite l’operato dei numerosi architetti e ingegneri, contribuì alla diffusione della concezione di architettura della difesa “alla moderna”, intendendo evidenziare con quali dinamiche i sistemi culturali e la tecnologia militare sviluppati in Italia si sono poi diffusi in tutto il continente europeo prima e nel mondo successivamente. Segue una seconda parte dove viene trattata l’acquisizione mediante sistemi laser e sensori digitali con una disamina sulla strumentazione sperimentale utilizzata e relativi dati ottenuti, quindi, dopo una rappresentazione di alcuni esempi di studio e analisi sul territorio nazionale, di gestione dei perimetri delle città fortificate nel secondo dopoguerra, si passa a una descrizione generale del valore e del significato del Patrimonio culturale in ambito nazionale e internazionale Nella terza e ultima parte della discussione, a partire dall’analisi del modello teorico desunto dagli allegati UNESCO, ne viene verificata l’applicazione del campo pratico e, attraverso i rilievi svolti sul campo coadiuvati da un attento studio, vengono evidenziati gli aspetti salienti tipici del caso preso in esame. Le indagini hanno preso in considerazione solo alcune porzioni delle mura, dove è stato ritenuto fossero più evidenti le difformità dal fronte delle mura, andando così a creare dei tipici casi di studio del problema della Buffer Zone. La rappresentazione grafica in questo percorso di ricerca è stato uno strumento fondamentale, capace di visualizzare, materializzare, dare volume e comprendere i disegni di progetto e di riprodurre, attraverso modelli digitali, gli elementi della fortificazione andando a creare, ove richiesto, sezioni dei singoli manufatti o ambientali come nel caso dello studio della buffer zone veronese. Con la proposta di istituzione della nuova zona cuscinetto “visibile” e la conseguente creazione di uno strumento operativo per la sua gestione, interrogabile nel tempo, in accordo con le indicazioni del Comitato del Patrimonio Mondiale, l’intento di questa ricerca è stato quello di cercare di rispondere a due delle principali questioni affrontate nel primo Piano di Gestione: promuovere uno sviluppo sostenibile del Centro Storico di Verona e salvaguardarne lo skyline. Il presente lavoro rappresenta una tappa all’interno di quell’articolato percorso conoscitivo volto alla comprensione delle principali architetture militari censite all’interno della città e sul territorio.

Die Kartierung und Lokalisierung eines mobilen Roboters in seiner Umgebung ist eine wichtige Voraussetzung für dessen Autonomie. In dieser Arbeit wird der Einsatz eines 3D-Laserscanners zur Erfüllung dieser Aufgaben untersucht. Durch die optimierte Anordnung eines rotierenden 2D-Laserscanners werden hochauflösende Bereiche vorgegeben. Zudem wird mit Hilfe von ICP die Kartierung und Lokalisierung im Stillstand durchgeführt. Bei der Betrachtung zur Verbesserung der Bewegungsschätzung wird auch eine Möglichkeit zur Lokalisierung während der Bewegung mit 3D-Scans vorgestellt. Die vorgestellten Algorithmen werden durch Experimente mit realer Hardware evaluiert.

The use of mobile bar code scanners is expanding to markets beyond popular manufacturing uses, such as healthcare, environmental testing and professional services. The successful interaction of users with mobile bar code scanners is of great importance from a business and technology perspective as well as from the user standpoint. Interaction problems associated with these devices may cause errors in data collection and affect job performance due to frustration, resulting in a potential impact on internal costs. The primary objective of this research was to develop design guidelines for the reengineering of keyboard designs for mobile bar code scanners. A secondary objective was to determine the effectiveness of current bar code scanner designs and how they can be related to other mobile technologies to develop a link across research areas. To accomplish these objectives, three different mobile bar code scanner keyboard designs were investigated: multiple-alphabetical, single-alphabetical and single split-Qwerty. A total of 42 subjects (18 females and 24 males) were recruited to participate in the experiment that was conducted to validate the research hypotheses. Time and accuracy data were recorded during the experiment and workload/subjective questionnaires were given to each participant following interaction with the different mobile bar code scanners. The following conclusions were reached based on the experimental results and are considered the major contributions of this research. First, an alphabetically laid out keyboard with multiple keys results in more time to find a character, a higher percentage of time spent typing incorrect lines, and a higher error rate than an alphabetically laid out keyboard with single keys. Perceived workload was also higher for a keyboard with multiple keys versus one with single keys. Second, no significant differences were found in the time to find a character, the percentage of time spent typing incorrect lines, and the error rate between keyboards with single keys, whether they are laid out alphabetically or in a split-Qwerty design. Finally, training a user on how to use a specific mobile bar code scanner keyboard layout often results in higher character rates, less time spent typing incorrect lines, and a lower error rate.
Graduation date: 2006

This thesis reports the work done towards design and fabrication of a versatile and low cost, frequency domain DOT (Diffuse Optical Tomography) Imager. A design which uses only a single fiber for the source and a single fiber bundle for the detector is reported. From near the source, to diametrically opposite to the source, the detected intensity of scattered light varies by three to four orders in magnitude, depending on the tissue/phantom absorption and scattering properties. The photo multiplier tube’s (PMT’s) gain is controlled to operate it in the linear range, thus increasing the dynamic range of detection. Increasing the dynamic range by multi channel data acquisition is also presented. Arresting the oscillations of a stepper using a negative torque braking method is also adopted in this application for increasing the speed of data acquisition. The finite element method (FEM) for obtaining photon density solution to the transport equation and the model based iterative image reconstruction (MPBIIR) algorithm are developed for verifying the experimental prototype. Simulation studies presented towards the end of this thesis work provide insight into the nature of measurements. The optical absorption reconstructed images from the simulation, verified the validity of implementation of the reconstruction method for further reconstructions from data gathered from the developed imager. A single iteration of MOBIIR to segment the region of interest (ROI) using an homogeneous measurement estimate is presented. Using the single iteration MOBIIR to obtain a relatively more accurate starting value for the optical absorption coefficient, and the reconstruction results for data obtained from tissue mimicking solid epoxy-resin phantom with a single in-homogeneity inclusion is also presented to demonstrate the imager prototype.

This thesis reports the work done towards design and fabrication of a versatile and low cost, frequency domain DOT (Diffuse Optical Tomography) Imager. A design which uses only a single fiber for the source and a single fiber bundle for the detector is reported. From near the source, to diametrically opposite to the source, the detected intensity of scattered light varies by three to four orders in magnitude, depending on the tissue/phantom absorption and scattering properties. The photo multiplier tube’s (PMT’s) gain is controlled to operate it in the linear range, thus increasing the dynamic range of detection. Increasing the dynamic range by multi channel data acquisition is also presented. Arresting the oscillations of a stepper using a negative torque braking method is also adopted in this application for increasing the speed of data acquisition. The finite element method (FEM) for obtaining photon density solution to the transport equation and the model based iterative image reconstruction (MPBIIR) algorithm are developed for verifying the experimental prototype. Simulation studies presented towards the end of this thesis work provide insight into the nature of measurements. The optical absorption reconstructed images from the simulation, verified the validity of implementation of the reconstruction method for further reconstructions from data gathered from the developed imager. A single iteration of MOBIIR to segment the region of interest (ROI) using an homogeneous measurement estimate is presented. Using the single iteration MOBIIR to obtain a relatively more accurate starting value for the optical absorption coefficient, and the reconstruction results for data obtained from tissue mimicking solid epoxy-resin phantom with a single in-homogeneity inclusion is also presented to demonstrate the imager prototype.