arXiv subjects

Paul Goyes-Peñafiel

Publications and source records attributed to Paul Goyes-Peñafiel.

4 recordsLinked to original sources

arXiv2026

Depth-Aware Implicit Neural Representation Priors for 3D Gravity Inversion

Gravimetry images subsurface density contrasts associated with geological structures, geothermal systems, and intrusive bodies. Recovering a three-dimensional density model from gravity observations is highly ill-posed because of its non-uniqueness, limited data coverage, and the attenuation of the gravity field with depth. Classical inversion methods rely on explicit regularization and parameter tuning, whereas supervised deep-learning approaches require representative gravity--density pairs that are rarely available. This paper proposes an unsupervised depth-aware implicit neural representation for 3D gravity inversion. The density volume is represented by multiple coordinate-based neural networks assigned to overlapping depth slabs and optimized directly from the observed gravity measurements through the sensitivity matrix. Slab-specific Fourier features, physics-based depth gains, and scheduled regularization provide structural priors without requiring labeled density models. Experiments on four synthetic scenarios show that the proposed method provides better overall performance in terms of RMSE, PSNR, and SSIM than the evaluated conventional and neural baselines. It also recovers more compact and spatially coherent density bodies, improves the separation of nearby anomalies, preserves internal structures, and reconstructs their vertical extent better. These results indicate that the proposed depth-aware formulation helps to mitigate the depth ambiguity inherent in gravity inversion. In the field experiment, where no ground-truth density model was available, the method produced compact, separated, and vertically coherent anomalies consistent with the observed gravity pattern.

cs.AI↗

arXiv2025

Poststack Seismic Data Preconditioning via Dynamic Guided Learning

Seismic data preconditioning is essential for subsurface interpretation. It enhances signal quality while attenuating noise, improving the accuracy of geophysical tasks that would otherwise be biased by noise. Although classical poststack seismic data enhancement methods can effectively reduce noise, they rely on predefined statistical distributions, which often fail to capture the complexity of seismic noise. On the other hand, deep learning methods offer an alternative but require large and diverse data sets. Typically, static databases are used for training, introducing domain bias, and limiting adaptability to new noise poststack patterns. This work proposes a novel two-process dynamic training method to overcome these limitations. Our method uses a dynamic database that continuously generates clean and noisy patches during training to guide the learning of a supervised enhancement network. This dynamic-guided learning workflow significantly improves generalization by introducing variability into the training data. In addition, we employ a domain adaptation via a neural style transfer strategy to address the potential challenge of encountering unknown noise domains caused by specific geological configurations. Experimental results demonstrate that our method outperforms state-of-the-art solutions on both synthetic and field data, within and outside the training domain, eliminating reliance on known statistical distributions and enhancing adaptability across diverse data sets of poststack data.

physics.geo-ph↗

arXiv2025

CDDIP: Constrained Diffusion-Driven Deep Image Prior for Seismic Image Reconstruction

Seismic data frequently exhibits missing traces, substantially affecting subsequent seismic processing and interpretation. Deep learning-based approaches have demonstrated significant advancements in reconstructing irregularly missing seismic data through supervised and unsupervised methods. Nonetheless, substantial challenges remain, such as generalization capacity and computation time cost during the inference. Our work introduces a reconstruction method that uses a pre-trained generative diffusion model for image synthesis and incorporates Deep Image Prior to enforce data consistency when reconstructing missing traces in seismic data. The proposed method has demonstrated strong robustness and high reconstruction capability of post-stack and pre-stack data with different levels of structural complexity, even in field and synthetic scenarios where test data were outside the training domain. This indicates that our method can handle the high geological variability of different exploration targets. Additionally, compared to other state-of-the-art seismic reconstruction methods using diffusion models. During inference, our approach reduces the number of sampling timesteps by up to 4x.

physics.geo-ph↗

arXiv2025

Physically Guided Deep Unsupervised Inversion for 1D Magnetotelluric Models

The global demand for unconventional energy sources such as geothermal energy and white hydrogen requires new exploration techniques for precise subsurface structure characterization and potential reservoir identification. The Magnetotelluric (MT) method is crucial for these tasks, providing critical information on the distribution of subsurface electrical resistivity at depths ranging from hundreds to thousands of meters. However, traditional iterative algorithm-based inversion methods require the adjustment of multiple parameters, demanding time-consuming and exhaustive tuning processes to achieve proper cost function minimization. Recent advances have incorporated deep learning algorithms for MT inversion, primarily based on supervised learning, and large labeled datasets are needed for training. This work utilizes TensorFlow operations to create a differentiable forward MT operator, leveraging its automatic differentiation capability. Moreover, instead of solving for the subsurface model directly, as classical algorithms perform, this paper presents a new deep unsupervised inversion algorithm guided by physics to estimate 1D MT models. Instead of using datasets with the observed data and their respective model as labels during training, our method employs a differentiable modeling operator that physically guides the cost function minimization, making the proposed method solely dependent on observed data. Therefore, the optimization algorithm updates the network weights to minimize the data misfit. We test the proposed method with field and synthetic data at different acquisition frequencies, demonstrating that the resistivity models obtained are more accurate than those calculated using other techniques.

physics.geo-ph↗