06 Fakultät Luft- und Raumfahrttechnik und Geodäsie
Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/7
Browse
29 results
Search Results
Item Open Access Forming a hybrid intelligence system by combining Active Learning and paid crowdsourcing for semantic 3D point cloud segmentation(2023) Kölle, Michael; Sörgel, Uwe (Prof. Dr.-Ing.)While in recent years tremendous advancements have been achieved in the development of supervised Machine Learning (ML) systems such as Convolutional Neural Networks (CNNs), still the most decisive factor for their performance is the quality of labeled training data from which the system is supposed to learn. This is why we advocate focusing more on methods to obtain such data, which we expect to be more sustainable than establishing ever new classifiers in the rapidly evolving ML field. In the geospatial domain, however, the generation process of training data for ML systems is still rather neglected in research, with typically experts ending up being occupied with such tedious labeling tasks. In our design of a system for the semantic interpretation of Airborne Laser Scanning (ALS) point clouds, we break with this convention and completely lift labeling obligations from experts. At the same time, human annotation is restricted to only those samples that actually justify manual inspection. This is accomplished by means of a hybrid intelligence system in which the machine, represented by an ML model, is actively and iteratively working together with the human component through Active Learning (AL), which acts as pointer to exactly such most decisive samples. Instead of having an expert label these samples, we propose to outsource this task to a large group of non-specialists, the crowd. But since it is rather unlikely that enough volunteers would participate in such crowdsourcing campaigns due to the tedious nature of labeling, we argue attracting workers by monetary incentives, i.e., we employ paid crowdsourcing. Relying on respective platforms, typically we have access to a vast pool of prospective workers, guaranteeing completion of jobs promptly. Thus, crowdworkers become human processing units that behave similarly to the electronic processing units of this hybrid intelligence system performing the tasks of the machine part. With respect to the latter, we do not only evaluate whether an AL-based pipeline works for the semantic segmentation of ALS point clouds, but also shed light on the question of why it works. As crucial components of our pipeline, we test and enhance different AL sampling strategies in conjunction with both a conventional feature-driven classifier as well as a data-driven CNN classification module. In this regard, we aim to select AL points in such a manner that samples are not only informative for the machine, but also feasible to be interpreted by non-experts. These theoretical formulations are verified by various experiments in which we replace the frequently assumed but highly unrealistic error-free oracle with simulated imperfect oracles we are always confronted with when working with humans. Furthermore, we find that the need for labeled data, which is already reduced through AL to a small fraction (typically ≪1 % of Passive Learning training points), can be even further minimized when we reuse information from a given source domain for the semantic enrichment of a specific target domain, i.e., we utilize AL as means for Domain Adaptation. As for the human component of our hybrid intelligence system, the special challenge we face is monetarily motivated workers with a wide variety of educational and cultural backgrounds as well as most different mindsets regarding the quality they are willing to deliver. Consequently, we are confronted with a great quality inhomogeneity in results received. Thus, when designing respective campaigns, special attention to quality control is required to be able to automatically reject submissions of low quality and to refine accepted contributions in the sense of the Wisdom of the Crowds principle. We further explore ways to support the crowd in labeling by experimenting with different data modalities (discretized point cloud vs. continuous textured 3D mesh surface), and also aim to shift the motivation from a purely extrinsic nature (i.e., payment) to a more intrinsic one, which we intend to trigger through gamification. Eventually, by casting these different concepts into the so-called CATEGORISE framework, we constitute the aspired hybrid intelligence system and employ it for the semantic enrichment of ALS point clouds of different characteristics, enabled through learning from the (paid) crowd.Item Open Access Evaluierung generalisierter Gebäudegrundrisse in großen Maßstäben(2012) Filippovska, Yevgeniya; Fritsch, Dieter (Prof. Dr.-Ing. habil.)Bei der Erzeugung von Karten werden die darzustellenden räumlichen Objekte in Abhängigkeit des angestrebten Maßstabs ausgewählt, verändert und so arrangiert, dass deren Form und Verteilung zu einem bestmöglichen Verständnis der räumlichen Gegebenheiten führt. Dabei weist die kartographische Abbildung unvermeidliche und zuweilen tiefgreifende geometrische Veränderungen im Vergleich zur Realität auf, welche durch eine übergeordnete Kontrollinstanz zu verifizieren und bewerten sind. Hierfür strebt man eine formalisierte Qualitätsbewertung der Ergebnisse an, so dass sich entsprechende Prozesse, bevorzugt mit Hilfe automatisierter Werkzeuge, umsetzen lassen. Obwohl die Lesbarkeit der Gesamtkomposition einer Karte das Ziel ist, muss die Qualitätsbewertung zuerst auf der untersten Generalisierungsebene, der sogenannten Mikroebene erfolgen, indem die Geometrie- bzw. die Formveränderungen von Einzelobjekten bemessen werden. Neben dem Straßennetz dienen den Kartennutzern häufig vor allem markante Gebäude als Orientierungshilfe, welche aus diesem Grund nicht allzu großen Veränderungen unterliegen dürfen. Im Rahmen dieser Arbeit werden daher Qualitätscharakteristiken aufgezeigt, welche auf dem direkten Vergleich zweier Gebäudegrundrisse – Original und generalisiert – basieren. Die vorliegende Arbeit beginnt mit einer theoretischen Einführung in das Thema der Qualität von Geodaten. Anschließend wird ein Wahrnehmungstest vorgestellt, welcher die Bewertung generalisierter Grundrisse durch menschliche Betrachter vornimmt. Versuche diese Wahrnehmungsprozesse mathematisch zu formalisieren wird als Ähnlichkeitsschätzung bezeichnet, deren Grundlagen darauffolgend dargelegt sind. In diesem Zusammenhang wird eine einheitliche Klassifizierung der Objektmerkmale basierend auf der zugrundeliegenden Berechnungsmethode vorgeschlagen. Ein Überblick über die bislang zur Qualitätsbewertung der Generalisierung gelaufenen Forschungsarbeiten und eine kritische Auseinandersetzung dazu runden den derzeitigen Kenntnisstand zum Themengebiet ab. Daran anschließend werden neue Charakteristiken zur Ähnlichkeitsanalyse vorgestellt, welche die 2D-Gebäudeobjekte unter den Aspekten der Kontur- und Flächentreue hin vergleichen. Da eine Zuordnung zwischen den Formelementen allgemein nicht zweifelsfrei feststellbar ist, werden die Objekte geometrisch gemäß der Standardisierung von Geodaten als Punktmengen betrachtet. Dies erlaubt es, die geometrischen Berechnungen fast ausschließlich auf den Standardoperatoren der Mengentheorie aufzusetzen und mit den topologischen Algorithmen der Graphentheorie zu kombinieren. Zur Bewertung der Konturtreue werden Charakteristiken auf Basis der objektbildenden Randmengen aufgezeigt, welche Aufschluss über die maximale Abweichung und den Anteil der Überlappung gibt. Die Flächentreue wird unter einem quantitativen und einem räumlichen Aspekt betrachtet, wobei eine Differenzierung zwischen den Elementen der Strukturveränderungen vorgenommen wird. Um die Aussagekraft und Praxistauglichkeit der vorgeschlagenen Charakteristiken zu überprüfen, wird eine Evaluierung von generalisierten Gebäudegrundrissen auf der Mikro- und Makroebene durchgeführt. Dabei spielt insbesondere auch die anschauliche Präsentation der Ergebnisse eine zentrale Rolle, so dass verschiedene Möglichkeiten zur Darstellung der einzelnen Charakteristiken bezüglich einer guten Diskriminierbarkeit der Qualitätsangaben im Fokus stehen. Die Analyse der Ergebnisse zeigt, dass alle vorgeschlagenen Charakteristiken aussagekräftig sind und eine vielseitige Beschreibung verschiedener Qualitätsaspekte der Generalisierung in deren Gesamtheit ermöglichen.Item Open Access Design and development of a calibration solution feasible for series production of cameras for video-based driver-assistant systems(2022) Nekouei Shahraki, Mehrdad; Haala, Norbert (apl. Prof. Dr.)In this study, we reviewed the current techniques and methods in photogrammetry - especially close-range photogrammetry - and focused on camera calibration. We reviewed the new evolving field of video-based driver-assistant systems, their requirements and their applications. Exclusively of fisheye cameras and a general omnidirectional projection, we extended an existing camera calibration model to address our needs and functionality requirements. These extensions enable us to use the camera calibration model in real-time embedded mobile systems with low processing power. We also introduced the free-function model as a flexible and advantageous model for camera distortion modelling. This is a new approach for modelling the overall image distortion together with the local lens distortions that are estimated using a standard model during the calibration process. Using free-function model on different lens designs, one can achieve good calibration accuracies by modelling the very local lens distortion taking benefit from the flexibility of this model. We introduced optimization strategies for recalculation and image rectification. These optimizations are also used to minimize the amount of required processing power and device memory. This brings many advantages to variety of computational platforms such as FPGAs, x86 and ARM processors, and makes it possible to benefit from variety of parallel-processing techniques. This model is capable of being used in runtime and is an ideal calibration model for using in variety of machine vision solutions. We also discussed several important requirements for accurate camera calibration that we later used in hardware test stand design phase. We designed and developed two different test stands in order to realize the specifications and geometrical features of multiple-view test-field-based camera calibration referred to as bundle-block calibration. One of their special geometrical characteristics is the uniform point distribution, which corresponds to the uniform motion. Such a point distribution is beneficial when using calibration models such as free-function model that enable us to model of local lens distortion with good accuracy and quality all over the image. A very important feature of this test stand is having the capability of performing camera/sensor alignment testing, a feature which is very important for testing the geometrical alignment of the internal mechanical elements of each camera. Using automated machines and algorithms in test stand calibration increased the stability and accuracy of the calibration and thus ensured the quality and speed of the calibration for cameras. These test stands are capable of performing automatic camera calibration, suitable for applications such as series-production of cameras. As an accuracy -and flexibility evaluation step for the free-function model, we tested the free-function calibration model on real-world data using a stereo camera with added large local distortions taking images from a front vehicle similar to the conditions where real-world use-cases are defined. By performing the camera calibration, we compared the calibration results and accuracy parameters of the free-function model to a conventional calibration model. Using these calibration results, we generated a set of disparity maps and compared their density and availability, especially on the areas where the local distortion was present. We used this test to compare the capabilities of the proposed model to conventional ones in real-wold situations where large optical distortions could be present that cannot be easily modelled with conventional calibration models. The higher modelling capability and accuracy of the free-function model will generally influence those functions that are using the information of the disparity map or the derived 3D information as part of their input data and potentially leads to the better functionality or even their availability if local distortions are present in the image. There are many more use-cases in photogrammetry and computer-vision where a higher calibration accuracy is beneficial on hardware such as low-cost optics where sometimes optical distortion are available that cannot easily be modelled with classical models. These use-cases could all benefit from the flexibility and modelling accuracy of the free-function model.Item Open Access Automatic model reconstruction of indoor Manhattan-world scenes from dense laser range data(2013) Budroni, Angela; Fritsch, Dieter (Prof. Dr.-Ing.)Three-dimensional modeling has always received a great deal of attention from computer graphics designers and with emphasis on existing urban scenarios it became an important topic for the photogrammetric community and architects as well. The generation of three-dimensional models of real objects requires both efficient techniques to acquire visual information about the object characteristics and robust methods to compute the mathematical models in which this information can be stored. Photogrammetric techniques for measuring object features recover three-dimensional object profiles from conventional intensity images. Active sensors based on laser measurements are able to directly deliver three-dimensional point coordinates of an object providing a fast and reliable description of its geometric characteristics. In order to transform laser range data into consistent object models, existing CAD software products establish a valid support to manual based approaches. However, the growing use of three-dimensional models in different field of applications brings into focus the need for automated methods for the generation of models. The goal of this thesis is the development of a new concept for the automatic computation of three-dimensional building models from laser data. The automatic modeling method aims at a reconstruction targeted on building interiors with an orthogonal layout. For this purpose, two aspects are considered: the extraction of all surfaces that enclose the interior volume and the computation of the floor plan. As a final result, the three-dimensional model integrates geometry and topology of the interior in terms of its boundary representation. The main idea underlying the automatic modeling is based on plane sweeping, a technique referable to the concept of sweep representation used in computer graphics to generate solid models. A data segmentation driven by the sweep and controlled by a hypothesis-and-test approach allows to assign each laser point to a surface of the building interior. At the next step of the algorithm, the floor plan is recovered by cell decomposition based on split and merge. For a successful generation of the model every activity of the reconstruction workflow should be taken into consideration. This includes the acquisition of the laser data, the registration of the point clouds, the computation of the model and the visualization of the results. The dissertation provides a full implementation of all activities of the automatic modeling pipeline. Besides, due to the high degree of automation, it aims at contributing to the dissemination of three-dimensional models in different areas and in particular in BIM processes for architecture applications.Item Open Access New methods for 3D reconstructions using high resolution satellite data(2021) Gong, Ke; Fritsch, Dieter (Prof. Dr.-Ing. habil. Prof. h.c.)Item Open Access Parameterfreies hierarchisches Graph-Clustering-Verfahren zur Interpretation raumbezogener Daten(2004) Anders, Karl-Heinrich; Fritsch, Dieter (Prof. Dr.-Ing.)Die Notwendigkeit der automatischen Interpretation und Analyse von räumlichen Daten wird heutzutage immer wichtiger, da eine stetige Zunahme der digitalen räumlichen Daten zu verzeichnen ist. Dies betrifft auf der einen Seite Rasterdaten wie auch auf der anderen Seite Vektordaten, welche überwiegend auf unterschiedlichen Landschaftsmodellen basieren. Differenzen zwischen diesen Landschaftsmodellen bestehen u.a. in den Objektarten, dem Grad der Generalisierung oder der geometrischen Genauigkeit der gespeicherten Landschaftsobjekte. Die interaktive Prozessierung und Analyse von großen Datenbeständen ist sehr zeitaufwendig und teuer. Speziell die manuelle Analyse räumlicher Daten zum Zwecke der Datenrevision wird in Zukunft das Limit der technischen Umsetzbarkeit erreichen, da moderne Anforderungen an die Laufendhaltung der Daten zu immer kürzeren Aktualisierungszyklen führen. Die automatische Interpretation digitaler Landschaftsmodelle setzt die Integration von Methoden des räumlichen Data Mining bzw. Knowledge Discovery in raumbezogenen Daten innerhalb von Geographischen Informationssystemen (GIS) voraus. Zunächst beschreiben wir einen Ansatz zur Generierung von 3D-Gebäuden, welche als Hypothese aus Katasterkarten abgleitet werden. Diese Vorgehensweise stellt ein Beispiel für die DLM-Interpretation auf der Grundlage eines spezifischen Modells dar und kann zur schnellen Generierung von groben 3D-Stadtmodellen oder als Vorabinformation zur bildgestützten 3D-Gebäuderekonstruktion verwendet werden. Des weiteren stellen wir detailliert einen Ansatz zur Ableitung von ATKIS-Daten aus ALK-Daten vor, welcher ein Beispiel für die DLM-Interpretation basierend auf einem generischen Modell der DLM-Basiselemente darstellt und zur automatischen Laufendhaltung der Daten dient. Beide Ansätze führen direkt zum grundsätzlichen Problem der Gruppierung von räumlichen Objekten, welches generell unter dem Begriff des Clusterns zusammengefasst wird. Man unterscheidet zwei Arten von Clusterverfahren: überwachte und unüberwachte Methoden. Unüberwachte Cluster- oder Lernverfahren können für den dritten genannten Fall der DLM-Interpretation verwendet werden und sind gut geeignet für die Modellgeneralisierung und die kartographische Generalisierung von DLM-Daten, falls die Methoden in der Lage sind, Cluster mit beliebiger Form zu erkennen. Die bisher existierenden Verfahren benötigen jedoch zumeist verschiedenste Kenntnisse als Voraussetzung, wie z.B. die Verteilungsfunktion der Daten oder Schrankenwerte für Ähnlichkeitsmessungen bzw. Abbruchkriterien. Zudem finden viele Clusterverfahren nur Gruppierungen mit konvexer Form und erkennen keine Löcher (z.B. Maximum-Likelihood-Methoden). Der Hauptteil dieser Arbeit widmet sich einem neu entwickelten, unüberwachten Clusterverfahren zur automatischen Interpretation von raumbezogenen Daten. Das Verfahren heißt Hierarchisches Parameterfreies Graph-CLustering (HPGCL) und dient zur Erkennung von Clustern beliebiger Form. Es benötigt weder Parameter wie z.B. Schrankenwerte noch Annahmen über die Verteilung der Daten oder die Anzahl der Cluster. Die Neuartigkeit des HPGCL-Algorithmus besteht auf der einen Seite in der Anwendung der Hierarchie von Nachbarschaftsgraphen zur Definition der Nachbarschaft eines Einzelobjekts oder eines Objektclusters in allgemeiner Art und Weise, sowie auf der anderen Seite in der Definition eines Entscheidungskriteriums zur Ähnlichkeitsbestimmung von Clustern, welches medianbasiert ist und ohne Angabe von Schwellwerten auskommt. Der Nächste-Nachbar-Graph, der Minimal Spannende Baum, der Relative Nachbarschaftsgraph, der Gabriel-Graph und die Delaunay-Triangulation kommen im HPGCL-Algorithmus zum Einsatz. Es wird aufgezeigt, dass die hierarchische Beziehung dieser Nachbarschaftsgraphen in einem natürlichen Generalisierungsprozess im Sinne einer grob-zu-fein-Segmentierung eines Datensatzes genutzt werden kann. Als weiterer Aspekt des HPGCL-Algorithmus kann die Tatsache genannt werden, dass im allgemeinen eine begrenzte Anzahl von Clustern größer eins gefunden wird. Im Gegensatz dazu benötigen andere hierarchische Clusterverfahren generell die Minimalanzahl der zu findenden Cluster als Parameter, da ohne Abbruchkriterium sonst alle Objekte des Datensatzes in einem einzigen großen Cluster vereinigt werden. Die Arbeit untersucht detailliert den Einfluss eines einzelnen Nachbarschaftsgraphen in der Hierarchie auf das Ergebnis des Clusterings, und es wird die Verwendbarkeit des HPGCL-Algorithmus auf der Grundlage von verschiedenen Datensatztypen evaluiert. Anhand zweier Datensätze werden die Ergebnisse des HPGCL-Verfahrens mit den Resultaten eines durch Testpersonen durchgeführten manuellen Clusterings verglichen.Item Open Access Mathematical methods for camera self-calibration in photogrammetry and computer vision(2013) Tang, Rongfu; Fritsch, Dieter (Prof. Dr.-Ing. habil.)Camera calibration is a central subject in photogrammetry and geometric computer vision. Self-calibration is a most flexible and highly useful technique, and it plays a significant role in camera automatic interior/exterior orientation and image-based reconstruction. This thesis study is to provide a mathematical, intensive and synthetic study on the camera self-calibration techniques in aerial photogrammetry, close range photogrammetry and computer vision. In aerial photogrammetry, many self-calibration additional parameters (APs) are used increasingly without evident mathematical or physical foundations, and moreover they may be highly correlated with other correction parameters. In close range photogrammetry, high correlations exist between different terms in the ‘standard’ Brown self-calibration model. The negative effects of those high correlations on self-calibration are not fully clear. While distortion compensation is essential in the photogrammetric self-calibration, geometric computer vision concerns auto-calibration (known as self-calibration as well) in calibrating the internal parameters, regardless of distortion and initial values of internal parameters. Although camera auto-calibration from N≥3 views has been studied extensively in the last decades, it remains quite a difficult problem so far. The mathematical principle of self-calibration models in photogrammetry is studied synthetically. It is pointed out that photogrammetric self-calibration (or building photogrammetric self-calibration models) can – to a large extent – be considered as a function approximation problem in mathematics. The unknown function of distortion can be approximated by a linear combination of specific mathematical basis functions. With algebraic polynomials being adopted, a whole family of Legendre self-calibration model is developed on the base of the orthogonal univariate Legendre polynomials. It is guaranteed by the Weierstrass theorem, that the distortion of any frame-format camera can be effectively calibrated by the Legendre model of proper degree. The Legendre model can be considered as a superior generalization of the historical polynomial models proposed by Ebner and Grün, to which the Legendre models of second and fourth orders should be preferred, respectively. However, from a mathemtical viewpoint, the algebraic polynomials are undesirable for self-calibration purpose due to high correlations between polynomial terms. These high correlations are exactly those occurring in the Brown model in close range photogrammetry. They are factually inherent in all self-calibration models using polynomial representation, independent of block geometry. According to the correlation analyses, a refined model of the in-plane distortion is proposed for close range camera calibration. After examining a number of mathematical basis functions, the Fourier series are suggested to be the theoretically optimal basis functions to build the self-calibration model in photogrammetry. Another family of Fourier self-calibration model is developed, whose mathematical foundations are the Laplace’s equation and the Fourier theorem. By considering the advantages and disvantages of the physical and the mathematical self-calibration models, it is recommended that the Legendre or the Fourier model should be combined with the radial distortion parameters in many calibration applications. A number of simulated and empirical tests are performed to evaluate the new self-calibration models. The airborne camera tests demonstrate that, both the Legendre and the Fourier self-calibration models are rigorous, flexible, generic and effective to calibrate the distortion of digital frame airborne cameras of large-, medium- and small-formats, mounted in single- and multi-head systems (including the DMC, DMC II, UltraCamX, UltraCamXp, DigiCAM cameras and so on). The advantages of the Fourier model result from the fact that it usually needs fewer APs and obtains more reliable distortion calibration. The tests in close range photogrammetry show that, although it is highly correlated with the decentering distortion parameters, the principal point can be reliably and precisely located in a self-calibration process under appropriate image configurations. The refined in-plane distortion model is advantageous in reducing correlations with the focal length and improving the calibration of it. The good performance of the combined “Radial + Legendre” and “Radial + Fourier” models is illustrated. In geometric computer vision, a new auto-calibration solution which needs image correspondences and zero (or known) skew parameter only is presented. This method is essentially based on the fundamental matrix and the three (dependent) constraints derived from the rank-2 essential matrix. The main virtues of this method are threefold. First, a recursive strategy is employed subsequently to a coordinate transformation. With an appropriate approximation, the recursion estimates the focal length and aspect ratio in advance and then calculates the principal point location. Second, the optimal geometric constraints are selected using error propagation analyses. Third, the final nonlinear optimization is performed on the four internal parameters via the Levenberg–Marquardt algorithm. This auto-calibration method is fast and efficient to obtain a unique calibration. Besides auto-calibration, a new idea is proposed to calibrate the focal length from two views without the knowledge of the principal point coordinates. Compared to the conventional two-view calibration techniques which have to know principal point shift a priori, this new analytical method is more flexible and more useful. Although the auto-calibration and the two-view calibration methods have not been fully mature yet, their good performance is demonstrated in both simulated and practical experiments. Discussions are made on future refinements. It is hoped that this thesis not only introduces the relevant mathematical principles into the practice of camera self-calibration, but is also helpful for the inter-communications between photogrammetry and geometric computer vision, which have many tasks and goals in common but simply using different mathematical tools.Item Open Access Integrated georeferencing for precise depth map generation exploiting multi-camera image sequences from mobile mapping(2020) Cavegn, Stefan; Haala, Norbert (apl. Prof. Dr.-Ing.)Image-based mobile mapping systems featuring multi-camera configurations allow for efficient geospatial data acquisition in both outdoor and indoor environments. We aim at accurate geospatial 3D image spaces consisting of collections of georeferenced multi-view RGB-D imagery, which may serve as basis for 3D street view services. In order to obtain high-quality depth maps, dense image matching exploiting multi-view image sequences captured with high redundancy needs to be performed. Since this process is entirely dependent on accurate image orientations, we mainly focus on pose estimation of multi-camera systems within this thesis. Nonetheless, we also present methods and investigations to obtain accurate, reliable and complete 3D scene representations based on multi-stereo mobile mapping sequences. Conventional image orientation approaches such as direct georeferencing enable absolute accuracies at the centimeter level in open areas with good GNSS coverage. However, GNSS conditions of street-based mobile mapping in urban canyons are often deteriorated by multipath effects and by shading of the signals caused by vegetation and large multi-story buildings. Moreover, indoor spaces do not even allow for any GNSS signals. Hence, we propose a powerful and versatile image orientation procedure that is able to cope with these issues encountered in challenging urban environments. Our integrated georeferencing approach extends the powerful structure-from-motion pipeline COLMAP with georeferencing capabilities. It assumes initial camera poses with sub-meter accuracy, which allow for direct triangulation of the complete scene. Such a global approach is much more efficient than an incremental structure-from-motion procedure. Furthermore, an initial image orientation solution already facilitates to georeference in a geodetic reference frame. Nevertheless, accuracies at the centimeter level can only be achieved by incorporation of ground control points. In order to obtain sub-pixel accurate relative orientations, strong tie point connections for the highly redundant multi-view image sequences are required. However, hardly overlapping fields of view, strongly varying views and weakly textured surfaces aggravate image feature matching. Hence, constraining relative orientation parameters among cameras is crucial for accurate, robust and efficient image orientation. Apart from supporting fixed multi-camera rigs, our integrated georeferencing approach that uses bundle adjustment allows for self-calibration of all relative orientation parameters or just single components. We extensively evaluated our integrated georeferencing procedure using six challenging real-world datasets in order to demonstrate its accuracy, robustness, efficiency and versatility. Four datasets were captured outdoors, one by a rail-based and three by different street-based multi-stereo camera systems. A portable mobile mapping system featuring a multi-head panorama camera collected two datasets in an indoor environment. Employing relative orientation constraints and ground control points within these indoor spaces resulted in absolute 3D accuracies of ca. 2 cm, and precisions at the millimeter level for relative 3D measurements. Depending on the use case, absolute 3D accuracy values for outdoor environments are slightly larger and amount to a few centimeters. However, determining 3D reference coordinates is a costly task. Not relying on any ground control points led to horizontal accuracies of ca. 5 cm for a scenario featuring some loops, while dropping down to a few decimeters for an extended junction area. Since the height component is even more dependent on prior camera poses from direct georeferencing, these 2D accuracies significantly decreased for the 3D case. However, incorporating just one ground control point facilitates the elimination of systematic effects, which results in 3D accuracies within the sub-decimeter range. Nevertheless, at least one additional check point is recommended in order to ensure a reliable solution. Once consistent and sub-pixel accurate relative poses of spatially adjacent images are available, in-sequence dense image matching can be performed. Aiming at precise and dense depth map generation, we evaluated several image matching configurations. Standard single stereo matching led to high accuracies, which could not significantly be improved by in-sequence matching. However, the image redundancy provided by additional epochs resulted in more complete and reliable depth maps.Item Open Access Crowd-sourced reconstruction of building interiors(München : Verlag der Bayerischen Akademie der Wissenschaften, 2016) Peter, Michael; Fritsch, Dieter (Prof. Dr.-Ing.)Location-based services (LBS) have gained huge commercial and scientific interest in recent years, due to the ubiquitous and free availability of maps, global positioning systems, and smartphones. To date, maps and positioning solutions are mostly only available for outdoor use. However, humans spend most of their time indoors, rendering indoor LBS interesting for applications such as location-based advertisement, customer tracking and customer flow analysis. Neither of the two prerequisites for indoor LBS - a map of the user's environment and a positioning system - is currently generally available: Most positioning methods currently under scientific investigation are based either on fingerprint maps of electro-magnetic signals (e.g. WiFi) or inertial measurement units. To overcome the flaws of these methods, they are often supported by models for the human movement which in turn rely on indoor maps. Ready-made maps, on the other hand, are generally unavailable due to indoor mapping being mostly manual, expensive and tedious. The vast amount of unmapped indoor space therefore calls for the transfer of methods used by Volunteered Geographic Information (VGI) communities like OpenStreetMap to indoor mapping. These methods comprise the digitization of features of interest such as building outlines from aerial images released to the community and the use of position traces. In this thesis, approaches are illustrated which can serve to enable this transfer. On the one hand, the thesis shows how photographs of evacuation plans - which are a compulsory part of the safety equipment of publicly used buildings in many countries - can substitute for the aerial images in the indoor domain. Due to the standardised nature of such plans, the manual digitization employed by VGI mappers in the outdoor domain can be replaced by an automatic reverse-engineering pipeline. To this end, the image is pre-processed and symbols, which depict evacuation routes or emergency equipment, are detected. Subsequently, foreground objects (i.e. walls) are distinguished from the background using an adequate binarisation operation. Based on the binary image, the sought-after vector information can be extracted by skeletonisation and skeleton tracing. The model is finalised by a bridging operation of the previously detected symbols which occlude parts of walls or stairs. As the model resulting from these operations is only available in a coordinate system defined by the original image, the transformation to a world-coordinate system or, at least, the unknown scale has to be determined. To this end, the indoor model is matched to an available model of the building's external shell. By detection of stairs, an approximate floor height can be computed and the 2D model is extruded to a 3D model. On the other hand, geometric features and semantic annotations may be added to existing models using pedestrian traces recorded by an indoor positioning system. As suitable generally available and low-cost systems do not exist yet, their existence is simulated in this work by a dead-reckoning system basing on a foot-mounted inertial measurement system. Methods for the derivation of the initial position and orientation necessary for the application of such a system are shown, as well as methods enabling the correction of remaining errors. The latter comprise an alignment approach using the external building shell and a map-matching method which employs the existing coarse model derived from the evacuation plan. Building on the collected pedestrian traces, semi-automatic and automatic approaches for the existing models' semantic and geometric refinement are presented which range from semantic annotation using the analysis of photographed doorplates to automatic door reconstruction. Furthermore, a geometric update of single rooms by conjoint analysis of the coarse model, pedestrian traces and a hand-held low-cost range camera is described. Lastly, works of indoor mapping are presented which are based on pedestrian traces and higher-level knowledge about the interior structure of the building modelled in an indoor grammar. Due to the differing characteristics of the two central elements of building interiors, corridors and rooms, the grammar is composed of a Lindenmayer system modelling the floor's corridor system and a split grammar describing the room layout which is found in the non-corridor spaces. The grammar is put to the test by applying it to distributedly collected noisy trace data.Item Open Access Automatic modeling of building interiors using low-cost sensor systems(2016) Khosravani, Ali Mohammad; Fritsch, Dieter (Prof. Dr.-Ing. habil.)Indoor reconstruction or 3D modeling of indoor scenes aims at representing the 3D shape of building interiors in terms of surfaces and volumes, using photographs, 3D point clouds or hypotheses. Due to advances in the range measurement sensors technology and vision algorithms, and at the same time an increased demand for indoor models by many applications, this topic of research has gained growing attention during the last years. The automation of the reconstruction process is still a challenge, due to the complexity of the data collection in indoor scenes, as well as geometrical modeling of arbitrary room shapes, specially if the data is noisy or incomplete. Available reconstruction approaches rely on either some level of user interaction, or making assumptions regarding the scene, in order to deal with the challenges. The presented work aims at increasing the automation level of the reconstruction task, while making fewer assumptions regarding the room shapes, even from the data collected by low-cost sensor systems subject to a high level of noise or occlusions. This is realized by employing topological corrections that assure a consistent and robust reconstruction. This study presents an automatic workflow consisting of two main phases. In the first phase, range data is collected using the affordable and accessible sensor system, Microsoft Kinect. The range data is registered based on features observed in the image space or 3D object space. A new complementary approach is presented to support the registration task in some cases where these registration approaches fail, due to the existence of insufficient visual and geometrical features. The approach is based on the user’s track information derived from an indoor positioning method, as well as an available coarse floor plan. In the second phase, 3D models are derived with a high level of details from the registered point clouds. The data is processed in 2D space (by projecting the points onto the ground plane), and the results are converted back to 3D by an extrusion (room height available from the point height histogram analysis). Data processing and modeling in 2D does not only simplify the reconstruction problem, but also allows for topological analysis using the graph theory. The performance of the presented reconstruction approach is demonstrated for the data derived from different sensors having different accuracies, as well as different room shapes and sizes. Finally, the study shows that the reconstructed models can be used to refine available coarse indoor models which are for instance derived from architectural drawings or floor plans. The refinement is performed by the fusion of the detailed models of individual rooms (reconstructed in a higher level of details by the new approach) to the coarse model. The model fusion also enables the reconstruction of gaps in the detailed model using a new learning-based approach. Moreover, the refinement process enables the detection of changes or details in the original plans, missing due to generalization purposes, or later renovations in the building interiors.
- «
- 1 (current)
- 2
- 3
- »