<p>Unmanned Aerial Vehicles (UAVs) stellen sich als sehr flexible und praxistaugliche Geräte auch für die Vermessung dar. Der Großteil der Systeme wird derzeit mit handelsüblichen Kameras betrieben &ndash; daraus abgeleitete Orthophotos oder Punktwolken werden vielfach für die Planung und Dokumentation verschiedener Anwendungen erfolgreich eingesetzt. Der Nachteil, der sich daraus ergibt, liegt zum einen in der sehr aufwendigen Berechnung von 3D-Punktwolken aus hochauflösenden Kamerabildern, zum anderen aber auch in der Unfähigkeit, Vegetation zu durchdringen und damit ein Geländemodell mit hoher Zuverlässigkeit zu erzeugen. Ebenfalls schwierig ist das Durchdringen von halbdurchlässigen Medien (z. B. Wasser). Abhilfe schaffen Multisensorsysteme, welche neben Kameras auch entsprechende Abstandsmodule integriert haben und somit aus mehreren Datenströmen einen Mehrwert generieren können. Die Integration unterschiedlicher Sensoren inkl. der notwendigen Verortung bringt aber auch einige Herausforderungen mit sich, u. a. eine komplexe und aufwendige Kalibration.</p><p>Das Fraunhofer-Institut für Physikalische Messtechnik IPM in Freiburg hat ein leichtgewichtiges und kompaktes Multisensorsystem entwickelt, welches aus folgenden Teilen besteht: schnelles laserbasiertes Abstandsmessmodul, zwei Farbkameras, inertiale Messeinheit (Inertial Measurement Unit, IMU) und Positionierungssystem (Global Navigation Satellite System, GNSS). Alle mechanischen Komponenten und das Gehäuse wurden soweit gewichtsoptimiert ausgeführt, dass ein Gesamtgewicht von knapp über 2 kg realisiert werden konnte. Die maximale Messdistanz des auf dem Pulslaufzeitverfahren basierenden Abstandsmessmoduls liegt bei 300 m (bei einer idealen Reflektivität der Oberfläche von 100 %). Die Präzision der Einzelpunktmessung (3 σ) liegt bei 15 mm.</p><p>Durch Integration von zwei RGB-Kameras kann nicht nur die erfasste Punktwolke entsprechend texturiert werden, sondern auch hochaufgelöste Orthophotos gerechnet werden. Durch diese zwei Datenströme (Punktwolke und Bilder) gelingt eine vollautomatisierte Auswertung der Daten mit Hilfe des maschinellen Lernens schnell und zuverlässig.</p><p>Der wissenschaftliche Beitrag wird zum einen die Entwicklungsschritte des Systems im Detail präsentieren, dann aber auch konkrete Ergebnisse aus der Anwendung vorstellen und eine Charakterisierung (Evaluierung) der Systemkomponenten darlegen.</p><p>&nbsp;</p>

<p>Für die Arbeiten im Gleisbereich ist ein System zur Erfassung von Zugfahrten unverzichtbar. Künstliche Intelligenz kann einen Teil dazu beitragen. Im Folgenden wird ein erster Ansatz, basierend auf Deep Learning Technologie, vorgestellt, welcher auf Grundlage von Bilddaten automatisch Züge detektiert und so zu mehr Sicherheit im Gleisbereich beitragen kann. Darüber hinaus ist eine Softwarelösung entwickelt worden, welche den umfangreichen und bisher arbeitsintensiven Datenvorbereitungsaufwand signifikant reduziert. Im vorliegenden Beitrag werden das Potenzial und die flexible Einsetzbarkeit von Künstlicher Intelligenz im ingenieurgeodätischen Kontext aufgezeigt.</p>

<p><span class="field-content">Im Jahresbericht werden die Projekte und Aktivitäten des i3mainz in komprimierter Form vorgestellt.</span></p>

<p>In the domain of computer vision, object recognition aims at detecting and classifying objects in data sets. Model-driven approaches are typically constrained through their focus on either a specific type of data, a context (indoor, outdoor) or a set of objects. Machine learning-based approaches are more flexible but also constrained as they need annotated data sets to train the learning process. That leads to problems when this data is not available through the specialty of the application field, like archaeology, for example. In order to overcome such constraints, we present a fully semantic-guided approach. The role of semantics is to express all relevant knowledge of the representation of the objects inside the data sets and of the algorithms which address this representation. In addition, the approach contains a learning stage since it adapts the processing according to the diversity of the objects and data characteristics. The semantic is expressed via an ontological model and uses standard web technology like SPARQL queries, providing great flexibility. The ontological model describes the object, the data and the algorithms. It allows the selection and execution of algorithms adapted to the data and objects dynamically. Similarly, processing results are dynamically classified and allow for enriching the ontological model using SPARQL construct queries. The semantic formulated through SPARQL also acts as a bridge between the knowledge contained within the ontological model and the processing branch, which executes algorithms. It provides the capability to adapt the sequence of algorithms to an individual state of the processing chain and makes the solution robust and flexible. The comparison of this approach with others on the same use case shows the efficiency and improvement this approach brings.</p>

<p><span style="color: rgb(73, 72, 72); font-family: Roboto, Roboto, &quot;Helvetica Neue&quot;, sans-serif; font-size: 14px; text-align: justify;">Experts&rsquo; knowledge about optical technologies for spatial and spectral recording is logically structured and stored in an ontology-based knowledge representation with the aim to provide objective recommendations for recording strategies. Besides operational functionalities and technical parameters such as measurement principles, instruments, and setups further factors such as the targeted application, data, physical characteristics of the object, and external influences are considered creating a holistic view on spectral and spatial recording strategies. Through this approach impacting factors on the technologies and generated data are identified. Semantic technologies allow to flexibly store this knowledge in a hierarchical class structure with dependencies, interrelations and description logic statements. Through an inference system the knowledge can be retrieved adapted to individual needs.</span></p>

<p>Accuracy, Artefact, Feature, Precision, Reconstruction, Resolution, Texture, Uncertainty are words central to many discussions of the documentation of CH. This chapter charts the interdisciplinary discussion towards a common understanding of terminologies used in CH. It is a discussion that recognizes critical differences or common misuse, and aims to contribute to a shared understanding that may be useful for all knowledge domains in the field.</p>

<p>Dieser Beitrag beschreibt die a priori Abschätzung erreichbarer Genauigkeiten der Bündelblockausgleichung. Dies geschieht mit Hilfe des EMVA1288 Standards. Durch numerische Simulationen wird hierzu zunächst die Zentrumsunsicherheit der Zielmarken mit der daraus folgenden Objektraum-Unsicherheit verknüpft. Der nächste Schritt ist eine Verknüpfung der EMVA1288-Kennzahlen und der daraus resultierenden Unsicherheit eines Grauwertes mit Algorithmen zur Ellipsendetektion. Abschließend wird ein stochastisches Modell vorgeschlagen und an einer real durchgeführten Kamerakalibrierung untersucht.</p>

<p>Application and analysis capabilities of 3D data sets generated through UAV photogrammetry are presented on the basis of an exemplary millstone quarry of the so-called Mayen quarry district (Germany). In general, millstone quarries are technical cultural heritage sites which give evidence for economic interpretations as they are of interest for the reconstruction of ancient procurement patterns. Particularly those quarries which had a high economic impact cover large areas with often arduous accessibility. Therefore, photogrammetric recording using a UAV is the best approach for fast and inexpensive high-quality documentation. The images are used to create a 3D data set and orthophoto. For analysis and interpretation the Spatial Image analysis and Visualisation Tool (SIVT) of i3mainz is used allowing a variety of interactive visualization functionalities. The 3D data are transformed into 2.5D data enabling segmentation of spatial information and volume calculations. Both functionalities support the cultural heritage expert&#39;s research: On the one hand the interactive segmentation allows producing a map of the quarry displaying only those parts associated with the extraction. On the other hand the output of the quarry including debris and millstone blanks can be calculated easily. All in all, the entire workflow beginning with data capture using a UAV followed by data processing, (2.5D / 3D) data analyses and visualisation of the results is presented.</p>

<p>Das Vorhaben will einen Innovationsschub für die Anwendung vermessungstechnischer<br />Tachymeter (auch Totalstationen, mobile Polarmesssysteme großer Reichweite)<br />auslösen.<br />Ermöglicht wird dieser Innovationsschub durch die Integration bildgebender Sensoren<br />mit einem konsequent modularen Lösungsansatz an Stelle der bislang mit nur<br />mäßigem Erfolg unternommenen monolithischen Lösungsansätze. Durch unser Konzept<br />werden die stark unterschiedlichen Entwicklungszyklen der zentralen Systemkomponenten<br />entkoppelt. Dadurch werden mittelfristig insbesondere KMU, Entwicklungseinrichtungen<br />und Hochschulen in die Lage versetzt, anwendungsspezifische<br />Lösungen für den expandierenden Markt von Geoinformationen und den steigenden<br />Bedarf an räumlichen Geometriedaten zu entwickeln und in der Folge auch eigenständig<br />zu vermarkten.</p>

<p>Traditionally stone inscriptions or drawings are documented through pictures or rubbings. The latter ones represent an analogue copy of the stone&rsquo;s surface and its features which are reproduced on paper. The disadvantage of this technique is the physical impact to the stone and the contained elements. Images reproduce the surface without contact. However, they might be affected by geometrical distortions and need appropriate lighting conditions to show the signs properly.</p><p>These problems will be avoided by means of non-contact 3D measuring techniques, like fringe projection. Such high resolution 3D techniques provide an exact geometrical copy of the original petroglyph, offering better results in legibility compared to traditional techniques. Moreover, it gives a more objective base for analysis and has less impact on the sometimes sensitive and eroded surfaces. Furthermore 3D data allows more extensive and further possibilities in processing and gives better preconditions for the interpretation.</p><p>However, depending on factors like resolution, scanned surface and degree of overlap between individual scans original 3D datasets may represent large up to really massive volumes of data. An effective use of such datasets can only be realised if they are condensed and prepared in a suitable way. This means reduction of the data volume, minimising any disturbing influence emerging from the spatial shape of the surface and emphasizing relevant information. The corresponding preparation of the data will then be a good base for a interpretation performed by the human science specialist through an adapted visualization. In addition the data should be prepared for high performance presentation to a wider community via the internet.</p><p>Processed digital copies of the Petroglyphs are visualised in order to enable the user to inspect the processed scans of the objects. By inspecting the scans the application provides a mass of functionality for achieving different views into the Petroglyphs and their appearance. This comprises on the one hand a simple 2D viewer for the processed data, and on the other hand a 3D viewer with interactive changeable light positions and water levels as well as a viewer for applying various lookup tables (colour), predefined image filters (convolution) and template matching (matching) regarding individual characters.</p><p>Provided functionality of the 3D viewer is based on features of 3D computer graphics. Surface normal vectors from the grey values of the processed scans and a light direction vector from an interactively changeable light source are computed. In addition shading is complemented by water filling, whereby the gray values are limited by the water level selected. Individual modifications are possible to improve the subjective impression by the user, trying to support him in his process of interpretation. Interactive changes of the light source directly affect the shading of the surface and provide a better idea of the 3D surface of the inscription board. Dynamic virtual water filling enables the user to obtain an even better impression of the depth of the individual characters and emphasise weathered characters.</p><p>The paper will explain the developed techniques and document its potential at selected data sets.</p>

<p>Dieser Beitrag befasst sich mit dem Einfluss der Zielmarkenexzentrizität während der Kamera-<br />Kalibrierung in Kombination mit verschiedenen Kalibrierkörpern. Zunächst wird<br />deren Einfluss auf die Resultate anhand numerischer Simulationen nachgewiesen. In diesen<br />Simulationen wird eine Erfassung des Einflusses der Exzentrizität auf Bildmessung, Objekt-<br />und Kamerageometrie angestrebt. Im zweiten Schritt wird eine Realkalibrierung mit<br />vergleichbarer Aufnahmekonfiguration durchgeführt. Dabei werden Gemeinsamkeiten bzw.<br />Unterschiede der erreichten Ergebnisse diskutiert sowie eine mögliche Kompensation des<br />Einflusses in der Praxis erörtert.</p>

<p><span class="Abstract" id="scm6MainContent_rptSections_lblSection_0">Modern instruments like laser scanner and 3D cameras or image based techniques like structure from motion produce huge point clouds as base for further object analysis. This has considerably changed the way of data compilation away from selective manually guided processes towards automatic and computer supported strategies. However it&rsquo;s still a long way to achieve the quality and robustness of manual processes as data sets are mostly very complex. Looking at existing strategies 3D data processing for object detections and reconstruction rely heavily on either data driven or model driven approaches. These approaches come with their limitation on depending highly on the nature of data and inability to handle any deviation. Furthermore, the lack of capabilities to integrate other data or information in between the processing steps further exposes their limitations. This restricts the approaches to be executed with strict predefined strategy and does not allow deviations when and if new unexpected situations arise. We propose a solution that induces intelligence in the processing activities through the usage of semantics. The solution binds the objects along with other related knowledge domains to the numerical processing to facilitate the detection of geometries and then uses experts&rsquo; inference rules to annotate them. The solution was tested within the prototypical application of the research project &ldquo;Wissensbasierte Detektion von Objekten in Punktwolken für Anwendungen im Ingenieurbereich (WiDOP)&rdquo;. The flexibility of the solution is demonstrated through two entirely different USE Case scenarios: Deutsche Bahn (German Railway System) for the outdoor scenarios and Fraport (Frankfort Airport) for the indoor scenarios. Apart from the difference in their environments, they provide different conditions, which the solution needs to consider. While locations of the objects in Fraport were previously known, that of DB were not known at the beginning. &copy; (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.</span></p>

<p>Constant technological progress results in new possibilities to produce reliable and rich spatial data of cultural heritage objects. In order to make optimal use of these capabilities, it is important to identify and name the information required to best serve the reasoning processes in these application fields. Correspondingly it is necessary to know about the characteristics of digitization techniques producing the content adapted to the needs of the applications. Due to the considerable complexity of instruments and processes producing the data, it is helpful to have a clear structure which relates the capabilities of the instruments to the requirements of the applications. This paper addresses this topic and shows a way of structuring spatial techniques as well as how this structure can be related to applications in the field of cultural heritage.</p>

<p>This paper documents the formulation of an international, interdisciplinary study, on a concerted European level, to prepare an innovative, reliable, independent and global knowledge base facilitating the use of today&rsquo;s and future optical measuring techniques for the documentation of&nbsp; cultural heritage. Cultural heritage professionals, color engineers and scientists share similar goals for the documentation, curation, long-term preservation and representation of cultural heritage artifacts. Their focus is on accuracy in the digital capture and remediation of artefacts through a range of temporal, spatial and technical constraints. A shared vocabulary to interrogate these shared concerns will transform mutual understanding and facilitate an agreed movement forward in cultural heritage documentation here proposed in the work of the COST Action Color and Space in Cultural Heritage (COSCH). The goal is a model that captures the shared concerns of professionals for a standards-based solution with an organic Linked Data model. The knowledge representation proposed here invokes a GUI interface for non-expert users of capture technologies, facilitates, and formulates their engagement with key questions for the field.</p>

<p>Results from two sample projects will be presented, showing possibilities to improve quality and precision of data capture and evaluation.</p><p>One example explains the potential to improve the absolute positional accuracy of a robot guided effector when applying photogrammetric strategies. Need and aims of such a tracking process are explained, followed by a description of actual solutions and their restrictions. Then the potential of photogrammetric solutions for such purposes are outlined, followed by a detailed description of the system developed. Finally, practical tests are shown, underlining that the design and realization are able to hold the challenging aims and improve the absolute accuracy of a robot by a factor of 20.</p><p>In an second example possibilities to evaluate huge data sets are presented. Due to the increasing availability of large unstructured point clouds obtained from laser scanning and/or photogrammetric data, there is a growing demand for automatic processing methods. Given the complexity of the underlying problems, several new methods resort to using semantic knowledge in particular for supporting object detection and classification. A novel approach making use of advanced algorithms is shown, in order to benefit from intelligent knowledge management strategies for the processing of 3D point clouds along with object classification in scanned scenes. In particular, these method extends the use of semantic knowledge to all stages of the processing, including the guidance of the 3D processing algorithms. The complete solution consists in a multi-stage iterative concept based on three factors: the modeled knowledge, the package of algorithms, and the classification engine.</p>

<p>The modeling of real-world scenarios through capturing 3D digital data has been proven applicable in a variety of industrial applications, ranging from security, to robotics and to fields in the medical sciences. These different scenarios, along with variable conditions, present a challenge in discovering flexible appropriate solutions. In this paper, we present a novel approach based on a human cognition model to guide processing. Our method turns traditional data-driven processing into a new strategy based on a semantic knowledge system. Robust and adaptive methods for object extraction and identification are modeled in a knowledge domain, which has been created by purely numerical strategies. The goal of the present work is to select and guide algorithms following adaptive and intelligent manners for detecting objects in point clouds. Results show that our approach succeeded in identifying the objects of interest while using various data types.</p>

<p><span style="line-height: 1.538em;">Due to the increasing availability of large unstructured point clouds obtained from laser scanning and/or photogrammetric data, there is a growing demand for automatic processing methods. Given the complexity of the underlying problems, several new methods try to using semantic knowledge in particular for supporting object detection and classification. In this paper, we present a novel approach which makes use of advanced algorithms to benefit from intelligent knowledge management strategies for the processing of 3D point clouds and for object classification in scanned scenes. In particular, our method extends the use of semantic knowledge to all stages of the processing, including the guidance of the 3D processing algorithms. The complete solution consists in a multi-stage iterative concept based on three factors: the modelled knowledge, the package of algorithms, and the classification engine. Two case studies illustrating our approach are presented in this paper. The studies were carried out on scans of the waiting area of an airport and along the tracks of a railway. In both cases the goal was to detect and identify objects within a defined area. With our results we demonstrate the applicability of our approach.</span></p><p>&nbsp;</p><p><span style="line-height: 1.538em;">Automatische Detektion und Klassifikation von Objekten in Punktwolken unter Nutzung mehrschichtiger Semantik. Infolge der zunehmenden Verfügbarkeit großer unstrukturierter Punktwolken aus Laserscanning und Photogrammetrie entsteht wachsender Bedarf für automatisierte Auswerteverfahren. Angesichts der häufig hohen Komplexität der in den Punktwolken enthaltenen Objekte stoßen rein datengetriebene Ansätze an ihre Grenzen. Es entstehen vermehrt Konzepte, die auf verschiedene Weise auch Gebrauch von der Semantik machen. Semantik und Algorithmik sind dabei oft eng miteinander verwoben und führen zu Limitationen in Art und Umfang der nutzbaren Semantik. Mit der vorgestellten Lösung werden Algorithmik und Semantik klar getrennt und mit den exakt auf diese Domänen zugeschnittenen Werkzeugen behandelt. Deren prozedurale Verknüpfung führt dann zu einem neuen Verarbeitungskonzept, das eine nach unserem Kenntnisstand bislang nicht erreichte Flexibilität und Vielseitigkeit in der Nutzung unterschiedlichster Semantiken besitzt und auch die Steuerung der Algorithmen integriert. Die iterative Gesamtlösung fußt auf drei Säulen, nämlich dem modellierten Wissen, dem Pool der Algorithmen und dem Identifikationsprozess. Erreichbare Resultate werden an zwei Beispielen dokumentiert. Ein Beispiel befasst sich mit der Analyse von Punktwolken aus dem Bereich der Lichtraumvermessung an Bahntrassen, das zweite mit Räumlichkeiten in einem Flughafen. In beiden Fällen müssen bestimmte Objektarten aufgefunden und klassifiziert werden.</span></p>

<p>&nbsp;</p><p>&nbsp;</p><p>&nbsp;</p><p>&nbsp;</p>

<p>Cette thèse s&#39;intéresse au recalage de données issues de capteurs 3D et multispectraux pour l&#39;étude du patrimoine.Lorsque l&#39;on étudie ce type d&#39;objet, il y a souvent peu de points saillants naturels entre ces jeux de données complémentaires. Par ailleurs, l&#39;utilisation de mires optiques est proscrite.Notre problème est donc de recaler des données multimodales sans points caractéristiques.Nous avons développé une méthode de recalage basé sur le suivi des systèmes d&#39;acquisition en utilisant des techniques issues de la photogrammétrie.Des simulations nous ont permis d&#39;évaluer la précision de la méthode dans trois configurations qui représentent des cas typiques dans l&#39;étude d&#39;objets du patrimoine.Ces simulations ont montré que l&#39;on peut atteindre une précision du suivi de 0.020 mm spatialement et 0.100 mrad angulairement en utilisant quatre caméras 5 Mpx lorsque l&#39;on numérise une zone de 400 mm x 700 mm.La précision finale du recalage repose sur le succès d&#39;une série de calibrations optiques et géométriques, ainsi que sur leur stabilité pour la durée du processus d&#39;acquisition.Plusieurs tests ont permis d&#39;évaluer la précision du suivi et du recalage de plusieurs jeux de données indépendants; d&#39;abord seulement 3D, puis 3D et multispecrales.Enfin, nous avons étendu notre méthode d&#39;estimation de la réflectance à partir des données multispectrales lorsque celles-ci sont recalées sur un modèle 3D.</p><p><br />The concern and interest of this PhD thesis is the registration of featureless 3D and multispectral datasets describing cultural heritage objects.In this context, there are few natural salient features between the complementary datasets, and the use of targets is generally proscribed.We thus develop a technique based on the photogrammetric tracking of the acquisition systems in use.A series of simulations was performed to evaluate the accuracy of our method in three configurations chosen to represent a variety of cultural heritage objects.These simulations show that we can achieve a spatial tracking accuracy of 0.020 mm and an angular accuracy of 0.100 mrad using four 5 Mpx cameras when digitizing an area of 400 mm x 700 mm. The accuracy of the final registration relies on the success of a series of optical and geometrical calibrations and their stability for the duration of the full acquisition process.The accuracy of the tracking and registration was extensively tested in laboratory settings. We first evaluated the potential for multiview 3D registration. Then, the method was used for to project of multispectral images on 3D models.Finally, we used the registered data to improve the reflectance estimation from the multispectral datasets&nbsp;</p>