OpendTect allows geoscientists to integrate and transform data from 1 D model to a predicted 3D rock property cube.

Integrated Machine Learning Rock Property Prediction Workflow

Our volume builder, a tool utilized to apply volume-based operations, enables users to build a low frequency model guided by a dense set of horizons and the interpolation of acoustic impedance log.
An inverted model-driven AI is outputted using Bayesian Linear Inversion tool. The inverted Acoustic Impedance can then be used as an input for Machine Learning to provide confidence estimates for rock properties prediction such as a porosity.

In this video, we demonstrate the power of data integration, the usage of modern data science and extreme gradient boosting algorithm to predict porosity in the F3 Dutch Offshore field. Results using state of the art techniques shows a good match with the blind well.