🤖 AI Summary
To address the challenges of digitizing, retrieving, and preserving 1.7 million historical aerial photographs—scattered across archival institutions in 65 countries during the 20th century—this study developed the first human-in-the-loop robotic scanning system. The system integrates industrial robot control, computer vision–driven precise film localization and quality assessment, adaptive image enhancement, and automated metadata annotation, enabling scalable, high-fidelity, low-damage digitization across diverse film formats. It achieves an average throughput of 240 frames per hour per workstation, with an error rate below 0.02% and a 30-fold increase in per-operator daily output. The project successfully completed high-quality digitization of the entire collection; a subset of the dataset has been publicly released for global academic use. This work significantly advances the accessibility, interoperability, and scholarly reusability of historical remote sensing data.
📝 Abstract
During the 20th Century, aerial surveys captured hundreds of millions of high-resolution photographs of the earth's surface. These images, the precursors to modern satellite imagery, represent an extraordinary visual record of the environmental and social upheavals of the 20th Century. However, most of these images currently languish in physical archives where retrieval is difficult and costly. Digitization could revolutionize access, but manual scanning is slow and expensive. Here, we describe and validate a novel robot-assisted pipeline that increases worker productivity in scanning 30-fold, applied at scale to digitize an archive of 1.7 million historical aerial photographs from 65 countries.