Subsurface Imaging

Subsurface imaging aims to reconstruct hidden structures and properties beneath the Earth's surface using various geophysical measurements. Current research emphasizes developing efficient and accurate methods, focusing on machine learning approaches like convolutional neural networks (CNNs), diffusion models, and physics-informed neural networks to improve the speed and accuracy of image reconstruction and prediction of subsurface phenomena, such as fluid flow and CO2 sequestration. These advancements are crucial for applications in carbon capture and storage, earthquake prediction, resource exploration, and environmental monitoring, offering improved safety and efficiency in these critical areas.