About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Publication
IPSJ 2024
Conference paper
Satellite Image Analysis by Pre-training Network
Abstract
Satellite image analysis is a crucial method for understanding climate change and vegetation growth. However, the relatively long acquisition periods of satellite images and the impact of weather conditions can make it challenging to determine accurate values related to vegetation and other factors. Therefore, this paper proposes a method that performs pre-training using a foundational model to convert synthetic aperture radar images into normalized difference vegetation index images, taking advantage of data from satellites other than the target satellite. Evaluation was conducted in a competition applied to vegetation understanding in a cabbage farm. As a result, our method achieved 1st place among 57 participants, with a 31% improvement in error compared to the 2nd-ranked method.