Céline Lacombe

Title: Céline Lacombe and Seismic Data Processing Using DUnet.

Introduction

Céline Lacombe is an inventor associated with Massy, France, and Cgg Services SAS. Her patent work concerns applying deep-learning methods to process seismic data and produce images of underground formations.

Latest Patents

Lacombe is associated with 1 patent, titled “Seismic Data Processing Using DUnet.” The invention describes a DUnet engine that processes seismic data using a contractive path to extract a code, an expansive path that reconstructs information with features passed through skip connections, and a model level that produces a processed image or an image showing the difference between the processed image and the input data.

The patent also describes selecting seismic data for training through an anchor method that expands an initial subset based on similarity measurements. A reweighting layer may combine inputs from different model layers to preserve signal-amplitude trends.

Career Highlights

Lacombe’s listed patent reflects work at the intersection of seismic imaging and neural-network-based data processing. The described approach aims to improve how seismic measurements are transformed into images for interpreting underground formations.

Collaborations

The listed coworkers associated with Lacombe include Hanyuan Peng and Jeremie Messud. The patent description outlines a technical approach involving multiple connected paths and training methods, though the provided information does not specify each coworker’s individual contribution.

Conclusion

Céline Lacombe’s patent work at Cgg Services SAS focuses on using DUnet-based processing to generate and refine seismic images, combining neural-network architecture with methods for training and preserving signal characteristics.