Abstract

Abstract. Close-range photogrammetry is an image-based technique which has often been used for the 3D documentation of heritage objects. Recently, advances in the field of image processing and UAVs (Unmanned Aerial Vehicles) have resulted in a renewed interest in this technique. However, commercially ready-to-use UAVs are often equipped with smaller sensors in order to minimize payload and the quality of the documentation is still an issue. In this research, two commercial UAVs (the Sensefly Albris and DJI Phantom 3 Professional) were setup to record the 19th century St-Pierre-le-Jeune church in Strasbourg, France. Several software solutions (commercial and open source) were used to compare both UAVs’ images in terms of calibration, accuracy of external orientation, as well as dense matching. Results show some instability in regards to the calibration of Phantom 3, while the Albris had issues regarding its aerotriangulation results. Despite these shortcomings, both UAVs succeeded in producing dense point clouds of up to a few centimeters in accuracy, which is largely sufficient for the purposes of a city 3D GIS (Geographical Information System). The acquisition of close range images using UAVs also provides greater LoD flexibility in processing. These advantages over other methods such as the TLS (Terrestrial Laser Scanning) or terrestrial close range photogrammetry can be exploited in order for these techniques to complement each other.

Highlights

  • Photogrammetry has long been employed for the documentation of heritage buildings and objects

  • Heritage documentation naturally benefits from these developments, as it complements terrestrial techniques (Nex and Remondino, 2014)

  • Combinations exist with other terrestrial techniques such as terrestrial laser scanners (TLS) (Barsanti et al, 2014; Grenzdörffer et al, 2015)

Read more

Summary

INTRODUCTION

Photogrammetry has long been employed for the documentation of heritage buildings and objects This technique enables the extraction of 3D information from 2D images, and is very useful in recording architectural or archeological details. Heritage documentation naturally benefits from these developments, as it complements terrestrial techniques (Nex and Remondino, 2014) Some examples in this domain include the modeling of façades (Cefalu et al, 2013; Fritsch et al, 2013a; Murtiyoso et al, 2016), whole buildings or monuments (Alidoost and Arefi, 2015; Chiabrando et al, 2015; Suwardhi et al, 2015), and post-catastrophe damage assessment (Achille et al, 2015; Baiocchi et al, 2013). Combinations exist with other terrestrial techniques such as terrestrial laser scanners (TLS) (Barsanti et al, 2014; Grenzdörffer et al, 2015)

Geometric Sensor Calibration
Dense Matching Approaches
TOOLS AND METHODS
Calibration
Aerotriangulation
Dense Matching
CONCLUSIONS
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call