In this paper, we present a development for recognizing objects from looted excavations. Experts with an archaeological background are not always available where an object needs to be assessed for tradability. For this purpose, we developed a smartphone app that can provide on-site assistance in the initial assessment of archaeological objects. The app sends captured images to a server for recognition and receives results with similar objects and their metadata along with an associated probability. A user can thus use these information to infer the provenance of the photographed object. To this end, a classifier was trained using a transfer learning procedure and the features of the trained network were used for an image matching procedure. The developed application will be tested by law enforcement agencies with a total of 15 smartphones for six months starting in early October.
Waldemar Berchtold, Huajian Liu, Simon Bugert, York Yannikos, Jingcun Wang, Julian Heeger, Martin Steinebach, Marco Frühwein, "Recognition of objects from looted excavations by smartphone app and deep learning" in Proc. IS&T Int’l. Symp. on Electronic Imaging: Mobile Devices and Multimedia: Enabling Technologies, Algorithms, and Applications, 2022, pp 362-1 - 362-4, https://doi.org/10.2352/EI.2022.34.3.MOBMU-362