NS artificial intelligence operated by NVIDIA Graphics cards help researchers remove clouds from satellite images. from researchers Osaka University showed that it is now possible to remove clouds from satellite images thanks to AI.
According to a study published in ScienceDirect, scientists from the university’s Division of Sustainable Energy and Environmental Engineering used “adversarial generative networks,” or GANs, with deep learning powered by NVIDIA graphics cards.
Method used to remove clouds
study author Kazunosuke Ikeno “By training the discriminating network to think that an image is real, we get reconstructed images that are more self-contained,” it said in a statement.
GANs are based on a pair of networks that compete with each other to produce realistic images. These competing networks allow developers to train artificial intelligence with limited data.
The first network, known as the “generative network”, reconstructs images without clouds. The second network, a “discriminatory network”, uses a convolutional neural network to select the images formed by the first network and real photographs.
The two competing networks augment each other without requiring a lot of data – the result is very realistic images without clouds.
Using 400-by-400-pixel images, the researchers trained the model to run on a PC Ubuntu Open-source operating system and a GeForce GTX 1060 GPU.
The method proposed in this study can be used in situations where buildings are hidden and visible by thin clouds. It can also be used in aerial photography to generate automatically monitored data sets for objects such as roads and rivers, as well as buildings viewed from street level.
.