Emerging Technologies for Quantitative Interpretation of Carbonate Rock Images

Explore emerging technologies for improving quantitative interpretations of carbonate rock images in this hour-long lecture. Delve into the application of machine learning and artificial neural networks for automatic identification of carbonate facies using the Dunham classification scheme. Learn about the use of high-resolution core images from the Integrated Ocean Discovery Program (IODP) Leg 194 to train a model using Python and TensorFlow. Discover how Convolutional Neural Networks (CNNs) can achieve up to 90% accuracy in identifying various carbonate textures, from Mudstone to Rudstone and Crystalline Dolomite. Compare the algorithm's performance to human geologists, noting similarities in biases and misclassifications. Understand the potential of these technologies to revolutionize descriptive data interpretation in the oil and gas industry, offering faster, more consistent, and high-resolution analysis. Gain insights into ongoing research aimed at expanding datasets, refining neural network training, and integrating image recognition with logging and petrophysical data estimation.

Emerging technologies to improve quantitative interpretations of carbonate rock images