Illegal excavation of archaeological sites aimed at collecting historical material culture (‘looting’) to introduce it in the illicit market of antiquities is a pressing problem on a global scale. Under favourable circumstances, looting can be exposed on Earth Observation (EO) data by detecting changes that have occurred between two or more consecutive EO images of a time-series (fig 1.). The main goal of ALCEO (Automatic Looting Classification from Earth Observation) project is to develop Artificial Intelligence (AI) methods for the automatic identification and classification of Cultural Heritage looted sites on EO multi-temporal series.
ALCEO aims to set a benchmark in the use of remote sensing for the identification of looting activities as: i) it will develop a novel and efficient semi-supervised change detection technique for identifying looting activities relying only on small number of labelled data ii) it will produce the first large EO training dataset of looted sites incorporating information provided by cultural heritage and Law Enforcement Agencies’ (LEA) experts; iii) it will develop new image restoration techniques to enhance the quality of EO images and make them more appropriate for looting detection tasks.
The proposed methodology for detection and tracking of looting activities will have a major impact on the protection of endangered cultural heritage sites by strengthening the ability of Law Enforcement Agencies to promptly react to ongoing illicit activities or acquire criminal conduct patterns to be used for behavioural profiling and further investigations.