A human-AI collaboration workflow for archaeological sites detection

This paper illustrates the results obtained by using pre-trained semantic segmentation deep learning models for the detection of archaeological sites within the Mesopotamian floodplains environment. The models were fine-tuned using openly available satellite imagery and vector shapes coming from a large corpus of annotations (i.e., surveyed sites). A randomized test showed that the best model reaches a detection accuracy in the neighborhood of 80%. Integrating domain expertise was crucial to define how to build the dataset and how to evaluate the predictions, since defining if a proposed mask counts as a prediction is very subjective. Furthermore, even an inaccurate prediction can be useful when put into context and interpreted by a trained archaeologist. Coming from these considerations we close the paper with a vision for a Human-AI collaboration workflow. Starting with an annotated dataset that is refined by the human expert we obtain a model whose predictions can either be combined to create a heatmap, to be overlaid on satellite and/or aerial imagery, or alternatively can be vectorized to make further analysis in a GIS software easier and automatic. In turn, the archaeologists can analyze the predictions, organize their onsite surveys, and refine the dataset with new, corrected, annotations.

Automated summary

This paper presents the results of using pre-trained semantic segmentation deep learning models to detect archaeological sites in the Mesopotamian floodplains environment. The models were fine-tuned using openly available satellite imagery and vector shapes. A randomized test showed that the best model achieved an accuracy of approximately 80%. The integration of domain expertise was crucial in constructing the dataset and evaluating the predictions, considering the subjective nature of determining if a proposed mask is a valid prediction. Additionally, even inaccurate predictions can be valuable when interpreted by trained archaeologists in the proper context. The paper concludes with a vision for a Human-AI collaboration workflow that combines annotated datasets, refined by human experts, with predictive models that can generate heatmaps overlaid on imagery or be vectorized for further analysis in GIS software, allowing archaeologists to analyze and refine the dataset with new annotations.