Test case on the quality analysis of structure from motion in airborne applications

For a few years, structure from motion has been commercially applied and 3D models are discussed in recent literature. Structure from motion is known to be a relatively fast and lowcost technique for the generation of photo-realistic 3D models. This technique uses a series of digital images of an object or site, taken from different positions. In this image series, each part of the object is recorded on at least three images. Combining these images with information from the metadata file allows to determine the different camera positions and orientations and to calculate a textured mesh or colored point set in 3D. These calculations are based on solving a system of geometric matrices and a least squares 2D to 3D projective problem. In this article, the geometric quality of an airborne digital surface model, based on structure from motion, is assessed, in order to use this model for archaeological research. An Iron Age hill fort in the municipality of Kooigem (Belgium) is used as test case for this research. In this area, archaeological features, such as walls and ditches, are known and documented. These objects have a submeter height difference with the surrounding topography and are therefore very useful for visual validation and geometrical analysis of the digital elevation models. Although this fast and low-cost method results in a fair visual representation of the environment of the relicts, a special focus is needed on less manifest objects of interest in the landscape. Besides, the geometrical accuracy of the final products will be discussed. A digital elevation model, acquired by airborne laser scanning, with an average point density of 4 points per square meter, is used for the statistical quality assessment of the elevation model generated by structure from motion. This statistical and visual analysis of the digital elevation model generated with structure from motion, allows to evaluate this new technique, its results and its applicability for archeological research.

[1]  Roberto Cipolla,et al.  Structure from motion , 2008 .

[2]  Frank Dellaert,et al.  Structure from motion without correspondence , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[3]  T. H. Kolbe,et al.  The Virtual 3D City Model of Berlin-Managing, Integrating, and Communicating Complex Urban Information , 2006 .

[4]  George Vosselman,et al.  Experimental comparison of filter algorithms for bare-Earth extraction from airborne laser scanning point clouds , 2004 .

[5]  Reinhard Koch,et al.  Automated reconstruction of 3D scenes from sequences of images , 2000 .

[6]  Jean Bourgeois,et al.  The use of stereoscopic images taken from a microdrone for the documentation of heritage – An example from the Tuekta burial mounds in the Russian Altay , 2011 .

[7]  3D BUILDING RECONSTRUCTION IMPROVEMENT BASED ON GIVEN GROUND PLAN INFORMATION AND SURFACE MODELS EXTRACTED FROM SPACEBORNE IMAGERY , 2012 .

[8]  Martin Fera,et al.  From Deposit to Point Cloud – a Study of Low-Cost Computer Vision Approaches for the Straightforward Documentation of Archaeological Excavations , 2011 .

[9]  Manolis I. A. Lourakis,et al.  SBA: A software package for generic sparse bundle adjustment , 2009, TOMS.

[10]  J. Hothmer.  Book reviewInternational archives of photogrammetry and remote sensing: ISPRS, Editor S. Murai: volume 27 part A, Tokyo-Japan-Japan 1989 , 1989 .

[11]  George Vosselman,et al.  Quality analysis on 3D building models reconstructed from airborne laser scanning data , 2011 .

[12]  B. Devereux,et al.  Visualisation of LiDAR terrain models for archaeological feature detection , 2008, Antiquity.

[13]  Birger Stichelbaut,et al.  Three-dimensional recording of archaeological remains in the Altai Mountains , 2012, Antiquity.

[14]  Rudi Goossens,et al.  3D building reconstruction based on given ground plan information and surface models extracted from spaceborne imagery , 2012 .

[15]  Fabio Remondino,et al.  Heritage Recording and 3D Modeling with Photogrammetry and 3D Scanning , 2011, Remote. Sens..

[16]  P. Maeyer,et al.  Kemmelberg (Belgium) case study : comparison of DTM analysis methods for the detection of relicts from the First World War , 2010 .

[17]  C. Heipke,et al.  AUTOMATIC PHOTOGRAMMETRIC PROCESSING OF SPOT IMAGERY FOR POINT DETERMINATION, DTM GENERATION AND ORTHOPROJECTION , 2010 .

[18]  R. Sliuzas,et al.  Spatio-temporal modelling of informal settlement development in Sancaktepe district, Istanbul, Turkey , 2011 .

[19]  Philippe De Maeyer,et al.  ACCURACY ASSESSMENT OF A LIDAR DIGITAL TERRAIN MODEL BY USING RTK GPS AND TOTAL STATION , 2011 .

[20]  Richard Szeliski,et al.  A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[21]  Markos Katsianis,et al.  A 3D digital workflow for archaeological intra-site research using GIS , 2008 .

[22]  A. Gruen.  ADAPTIVE LEAST SQUARES CORRELATION: A POWERFUL IMAGE MATCHING TECHNIQUE , 1985 .

[23]  Geert Verhoeven,et al.  Taking computer vision aloft – archaeological three‐dimensional reconstructions from aerial photographs with photoscan , 2011 .