Archiving 4.0: Application of Image Processing and Machine Learning for the Robben Island Mayibuye Archives

A groundbreaking study identifies the most effective machine learning models for face detection in historical archives, with significant implications for digitizing and analyzing visual records. Using a subset of images from the Robben Island Mayibuye Archives—a collection documenting South Africa's anti-apartheid struggle—the research evaluates multiple models based on metrics like Intersection over Union (IoU) and Average Precision, alongside their processing speed. Findings reveal that certain models excel in accuracy while others prioritize efficiency, offering tailored solutions for large-scale archival projects. This work advances Archiving 4.0, a broader initiative leveraging fourth industrial revolution (4IR) technologies to enhance archival practices in South Africa.

Kern, D., Zweng, M., Sello, S., Bagula, A., & Klauck, U. (2020). Archiving 4.0: Application of Image Processing and Machine Learning for the Robben Island Mayibuye Archives. 2020 International SAUPEC/RobMech/PRASA Conference, 1–6. https://doi.org/10.1109/saupec/robmech/prasa48453.2020.9041115