Colorization of Day-Night Aerial Infrared Images Using Online GIS Imagery

Humans find it difficult to recognize scenes within infrared images, which lack color and contrast. A common workaround is to convert infrared images into visible color images. Recent advances in deep learning have substantially enhanced image colorization techniques. Ideally, simultaneously captured visible and thermal image pairs are used for training. However, acquiring such image pairs is quite challenging under low-light conditions, such as nighttime. To address this challenge, we propose a novel colorization method using Google Earth. Using GPS and camera orientation data, Google Earth can provide visible images with the same field of view as infrared images. We mitigate the differences in capture timing and angles between the infrared and visible images using a two-step approach: image colorization and image fusion. For image colorization, we employ a CycleGAN architecture with content loss to preserve spatial structure while learning color mapping from visible reference images. For image fusion, we apply Multiscale Decomposition to extract and integrate fine structural details from the infrared images. Experiments using SWIR and MWIR infrared images demonstrate successful colorization in daytime as well as nighttime scenarios, confirming the effectiveness of our proposed method.