The BigImage toolkit available in GeoSlicer is a set of tools designed to make it easier to manipulate huge volumes outside of RAM. This allows the user to work with massive images of very high resolution, i.e. that exceed the amount of memory available on the machine. The challenge is to give the user enough […]
We recently performed a high resolution facies characterization study using our plugins developed for OpendTect by dGB Earth Sciences. This work was done in collaboration with our consulting partner Ricardo Zavala-Torres, Agustín Heriberto Domínguez Mendoza from PEMEX and Armando Fragoso from GEOKRONIX. The workflow includes:✅ Seismic Spectral Blueing✅ Wavelet Estimation✅ Low-Frequency Modeling✅ Bayesian Linearized Inversion✅ […]
4D Seismic monitoring provides crucial insights for reservoir management, but its high dimensionality and processing complexity significantly impact the efficiency of inversion problems like history matching with Ensemble Smoother with Multiple Data Assimilation (ES-MDA). 🔍 The Challenge:4D seismic data requires intensive processing, which can degrade assimilation performance and slow down model updates. 💡 Our Solution:By […]
This week, part of the LTrace team participated in a workshop on seismic inversion at the Universidade Federal de Santa Catarina, Connectionism and Cognitive Sciences Lab. We had the pleasure of welcoming Professor Dr. Dario Grana from the University of Wyoming as a special speaker. He delivered an insightful lecture on his consortium focused on […]
📢 Generated digital rock model with 3344×1360×8304 voxels 📢 We are thrilled to share our latest results from the MUSA project, where we integrate microCT and coreCT data using a novel multiscale approach!Based on CSinGAN (https://lnkd.in/dgx4D-WY), a generative model, we introduced the coreCT conditioner only at the generation stage. This ensures that unresolved phases in […]
Today, LTrace conducted a short course for Petrobras on the Thin Section environment of the GeoSlicer software. Our mentor for the course was Márcio Valério Weck Pereira Among its many capabilities, we applied our AI-driven segmentation algorithm for thin sections, enabling the identification of both minerals and textures (GIF below). 🧠💡 🔹 Reminder: GeoSlicer is […]
We’re happy to share the productive year LTrace had in 2024. Check out the contributions our team made across scientific journals and conferences:
ai, deeplearing, digitalrock, machine learning, news, reservoir engineering
📢 Modeling fluid flow in porous media with broad pore size distributions remains an open challenge, as balancing field of view and voxel resolution often requires difficult compromises. 🔬 Our latest developments 🔬We’re excited to share our developments on multiscale pore network models, implemented in the open-source solution Geoslicer. Multiscale PNM addresses this challenge by representing subresolution pores through effective properties while reducing the computational […]
At the end of last year LTrace was selected at a contest promoted by PETROBRAS and SEBRAE called “Petrobras connections for innovation – Startup module”
ai, history matching, kalman filter, machine learning, neural networks, ocean plugin, petrel, R&D, reservoir engineering, seismic
Join us at the SEG conference during the week of 15-20 September. We will be presenting our new developments using deep learning for rock properties prediction from micro CT rock images.