A GAN-Enhanced Deep Learning Framework for Rooftop Detection from Historical Aerial Imagery
It addresses a domain-specific challenge for urban development researchers by enabling more reliable extraction of building footprints from archival photographs, though it is incremental as it combines existing GAN methods.
This research tackled the problem of detecting rooftops from historical black-and-white aerial imagery by using a GAN-based two-stage image enhancement pipeline (colorization and super-resolution) to improve detection performance, resulting in YOLOv11n achieving over 85% mAP, a 40% improvement over original images.
Precise detection of rooftops from historical aerial imagery is essential for analyzing long-term urban development and human settlement patterns. Nonetheless, black-and-white analog photographs present considerable challenges for modern object detection frameworks due to their limited spatial resolution, absence of color information, and archival degradation. To address these challenges, this research introduces a two-stage image enhancement pipeline based on Generative Adversarial Networks (GANs): image colorization utilizing DeOldify, followed by super-resolution enhancement with Real-ESRGAN. The enhanced images were subsequently employed to train and evaluate rooftop detection models, including Faster R-CNN, DETReg, and YOLOv11n. The results demonstrate that the combination of colorization with super-resolution significantly enhances detection performance, with YOLOv11n achieving a mean Average Precision (mAP) exceeding 85\%. This signifies an enhancement of approximately 40\% over the original black-and-white images and 20\% over images enhanced solely through colorization. The proposed method effectively bridges the gap between archival imagery and contemporary deep learning techniques, facilitating more reliable extraction of building footprints from historical aerial photographs. Code and resources for reproducing our results are publicly available at \href{https://github.com/Pengyu-gis/Historical-Aerial-Photos}{github.com/Pengyu-gis/Historical-Aerial-Photos}.