Artificial Intelligence, 3D Documentation, and Rock Art—Approaching and Reflecting on the Automation of Identification and Classification of Rock Art Images

Researchers developed AI models capable of automatically identifying and classifying Bronze Age rock carvings (petroglyphs) with high accuracy, particularly for boats, circles, human figures, and wheels. Using a novel 3D documentation method that reduces bias, the team trained faster region-based convolutional neural networks (Faster-RCNN) on over 4000 images from 408 scans of rock art sites in Tanum, Sweden. Data augmentation and transfer learning enhanced model performance, demonstrating the potential for AI to unlock new insights into Bronze Age social structures and ideologies. This interdisciplinary study also reflects on the implications of integrating artificial intelligence with archaeology and digital humanities.

Horn, C., Ivarsson, O., Lindhé, C., Potter, R., Green, A., & Ling, J. (2021). Artificial Intelligence, 3D Documentation, and Rock Art—Approaching and Reflecting on the Automation of Identification and Classification of Rock Art Images. Journal of Archaeological Method and Theory, 29(1), 188–213. https://doi.org/10.1007/s10816-021-09518-6