Processing of photogrammetric datasets. A metrological viewpoint
25 April 2017
Ewelina Rupnik, Postdoctoral Fellow at IGN (Institut national de l’information géographique et forestière)
Today, photogrammetry is a mature technology, able to automatically process relatively large, standard datasets. The classical processing, however,
– does not necessarily furnish 3D products of high metrological quality,
– fails in certain imaging configurations (e.g. linear photogrammetry, low contrast images, repetitive scenery), and
– cannot handle images captured by sensors of different spectral sensitivity.
In this talk I will discuss my current research activities and short-term research perspectives, adressing some of the above mentioned challenges. Two real-case studies will be discussed:
– a case adapted to creation of high quality Digital Surface Models (DSM) from very high resolution (VHR) satellite imagery, and
– a case adapted to thermal imagery processing.