Enhancing and upscaling historic analog aerial images using artificial intelligence models

Basarić, Miloš and Potić, Ivan and Sekulović, Dragoljub (2023) Enhancing and upscaling historic analog aerial images using artificial intelligence models. In: Proceedings, SYM-OP-IS 2023 - 50th International Symposium on Operational Research. Medija centar “Odbrana”, Belgrade, pp. 217-220. ISBN 978-86-335-0836-0

Text 2023_Basaric-Potic-Sekulovic-Istorijski-analogni-snimci_Symopis_M63.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (3MB)

Abstract

This paper deals with the application of artificial intelligence techniques for the improvement of old aerial photogrammetric images obtained by film scanning. These images differ in the quality and resolution of the scans, but also in the state of film preservation, which presents challenges in the interpretation of the content. The study focuses on the use of deep learning-based Super-resolution reconstruction (SRR), a technique that generates high-resolution images from lower-resolution ones. The paper uses an online platform called Remini for image enhancement based on artificial intelligence (hereinafter referred to as AI) for the processing of these images, which results in improved detail contrast and reduced noise in the image. Improved images are of higher quality with greater and faster interpretation of content. The software's scalability allows for batch processing, making it suitable for large volumes of recordings. If the initial images are georeferenced, the spatial reference can be maintained with improved quality by reducing the dimensions of the resulting images to the dimensions of the initial images. The research highlights the practical value of using AI solutions to improve existing recordings and suggests further research for local implementation of AI models in this area.

Actions (login required)

View Item