Call for Paper

Apostolos SARRIS (1) | Melda KÜÇÜKDEMİRCİ (2) | Tuna KALAYCI (3)
((1) University of Cyprus and GeoSat Research Lab, Institute for Mediterranean Studies-FORTH, Cyprus / Greece | (2) Istanbul University Cerrahpaşa, Turkey | (3) Faculty of Archaeology, Leiden University, The Netherlands)

Keywords: Articificial Intelligence, Machine Learning, Remote Sensing, Satellite, Geophysics, Aerial

Call: The recent years are experiencing an increasing number of Artificial Intelligence/Deep Learning/Machine Learning (AI/DL/ML) applications in various domains of Digital Humanities. Still, we are in the infant stage of their application and there are a number of limitations faced mainly due to the lack of a large volume of data that are required for their successful implementation.
Thanks to the development of UAVs, Lidar, Hyper-spectral and multispectral satellite imaging, multi-sensor and motorized equipment for geophysical surveys, a large quantity of data that cover extensive areas of the landscapes with high resolution is created, which can only be analysed efficiently by automated approaches leading towards a fast and accurate interpretation. This session is looking to build on the experience obtained from the application of various algorithms of AI/DL/ML on remote sensing data, spanning from satellite, aerial, ground based/geophysical and underwater surveys. We want to provide an exchange platform of the experiences acquired in image pre-processing, pixel-based classification, feature recognition, segmentation, and feature extraction and scene understanding in relation to the AI/DL/ML-based algorithms: comparison of networks, pre-trained models, hyper-parameter choices, APIs and in addition to these, discussion on obstacles, limitations, challenges, key bottlenecks and potential future direction of these new methodologies.