Automatic detection and qualification of objects in point clouds using multi-layered semantics

Summary: Due to the increasing availability of large unstructured point clouds from lasers canning and photogrammetry, there is a growing demand for automatic evaluation methods. Given the complexity of the underlying problems, several new methods resort to using semantic knowledge in particular for object detection and qualification support. In this paper, we present a novel approach which makes use of advanced algorithms, and benefits from intelligent knowledge management strategies for the processing of 3D point clouds and object qualification in a scanned scene. 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 of a multi- stage, iterative, concept based on three factors: the modeled knowledge, the package of algorithms, and the qualification engine. Zusammenfassung: Automatische Detektion und Qualifizierung von Objekten in Punktwolken unter Nutzung mehrschichtiger Semantik. Infolge der zunehmenden Verfugbarkeit groser unstrukturierter Punktwolken aus Laserscanning und Photogrammetrie entsteht wachsender Bedarf fur automatisierte Auswerteverfahren. Angesichts der haufig hohen Komplexitat der in den Punktwolken enthaltenen Objekte stosen rein Daten-getriebene Ansatze an ihre Grenzen. Es entstehen vermehrt Konzepte, die auf verschiedene Weise auch Gebrauch von Semantik machen. Semantik und Algorithmik sind dabei oft eng miteinander verwoben und fuhren zu Limitationen in Art und Umfang der nutzbaren Semantik. Mit der vorgestellten Losung werden Algorithmik und Semantik klar getrennt und mit den exakt auf diese Domanen zugeschnittenen Werkzeugen behandelt. Deren prozedurale Verknupfung fuhrt dann zu einem neuen Verarbeitungskonzept, das eine bislang nicht erreichte Flexibilitat und Vielseitigkeit in der Nutzung unterschiedlichster Semantiken besitzt und auch die Steuerung der Algorithmen integriert. Die iterative Gesamtlosung fust auf drei Saulen, die aus dem modellierten Wissen, dem Pool der Algorithmen und dem Identifikationsprozess bestehen.

[1]  E. Rank,et al.  Towards a 3D Spatial Query Language for Building Information Models , 2006 .

[2]  Christophe Cruz,et al.  From 3D Point Clouds to Semantic Objects - An Ontology-based Detection Approach , 2011, KEOD.

[3]  Wenzhong Shi,et al.  Topology for 3D spatial objects , 2003 .

[4]  Luc Van Gool,et al.  A Probabilistic Approach to Building Roof Reconstruction Using Semantic Labelling , 2002, DAGM-Symposium.

[5]  Wolfram Burgard,et al.  Instace-Based AMN Classification for Improved Object Recognition in 2D and 3D Laser Range Data , 2007, IJCAI.

[6]  Christophe Cruz,et al.  Managing Semantics Knowledge for 3D Architectural Reconstruction of Building Objects , 2010, 2010 Eighth ACIS International Conference on Software Engineering Research, Management and Applications.

[7]  Monique Thonnat,et al.  Ontology based complex object recognition , 2008, Image Vis. Comput..

[8]  Germain Forestier,et al.  Ontology-Based Object Recognition for Remote Sensing Image Interpretation , 2007, 19th IEEE International Conference on Tools with Artificial Intelligence(ICTAI 2007).

[9]  Ben Taskar,et al.  Discriminative learning of Markov random fields for segmentation of 3D scan data , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[10]  Christophe Cruz,et al.  From quantitative spatial operator to qualitative spatial relation using Constructive Solid Geometry, logic rules and optimized 9-IM model: A semantic based approach , 2012, 2012 IEEE International Conference on Computer Science and Automation Engineering (CSAE).

[11]  Ian Horrocks,et al.  OWL Web Ontology Language Reference-W3C Recommen-dation , 2004 .

[12]  George Vosselman,et al.  Automatic extraction of building features from terrestrial laser scanning , 2006 .

[13]  Joachim Hertzberg,et al.  3D Mapping with Semantic Knowledge , 2005, RoboCup.

[14]  L. Stein,et al.  OWL Web Ontology Language - Reference , 2004 .

[15]  Ernst Rank,et al.  Implementing metric operators of a Spatial Query Language for 3D Building Models: Octree and B-Rep approaches , 2009 .

[16]  Nikos Paragios,et al.  Segmentation of building facades using procedural shape priors , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[17]  George Vosselman,et al.  3D BUILDING MODEL RECONSTRUCTION FROM POINT CLOUDS AND GROUND PLANS , 2001 .

[18]  Ernst Rank,et al.  Specification and implementation of directional operators in a 3D spatial query language for building information models , 2009, Adv. Eng. Informatics.

[19]  H. Lan,et al.  SWRL : A semantic Web rule language combining OWL and ruleML , 2004 .

[20]  Robert B. Fisher,et al.  Quality Enhancement of Reconstructed 3D Models Using Coplanarity and Constraints , 2002, DAGM-Symposium.

[21]  E. Rank,et al.  Topological operators in a 3D Spatial Query Language for Building Information Models , 2008 .

[22]  George Vosselman,et al.  Knowledge based reconstruction of building models from terrestrial laser scanning data , 2009 .

[23]  Shi Pu,et al.  AUTOMATIC EXTRACTION OF BUILDING FEATURES FROM TERRESTRIAL LASER SCANNING , 2006 .

[24]  Claus Brenner,et al.  Reconstruction of Façade Structures Using a Formal Grammar and RjMCMC , 2006, DAGM-Symposium.