Laura Galvis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Laura Galvis has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Henry Arguello
Universidad Industrial de Santander
9 shared publications Henry Arguello
Department of Computer Science, Universidad Industrial de Santander, Bucaramanga 680002, Colombia
5 shared publicationsGonzalo R. Arce
Department of Electrical and Computer Engineering, University of Delaware, Newark, Delaware 19716, USA
4 shared publicationsDaniel Lau
Department of Electrical and Computer Engineering, University of Kentucky, Lexington, Kentucky 40507, USA
2 shared publications- Jorge Bacca
Department of Computer Science, Universidad Industrial de Santander, Bucaramanga, Colombia
2 shared publications - Luis Rodríguez-López
Department of Computer Science, Universidad Industrial de Santander, Santander, Colombia
2 shared publications
Latest Publications
- 30 September 2026
- Published by Society of Exploration Geophysicists in The Leading Edge
Guided Diffusion Models for Seismic Data Reconstruction via Deep Image Priors and Restoration Operators
Seismic data are widely used in subsurface energy exploration. However, several factors affect seismic data quality. Seismic data reconstruction is essential when acquisition is incomplete because missing sources or receivers produce subsampled shot-gathers with incomplete trace coverage that degrades data quality and affects subsequent seismic processing. Various deep-learning methodologies, particularly generative models, have been developed to reconstruct seismic signals. In this context, diffusion models (DMs) have emerged for seismic data processing, either by guiding image generation or by guiding the reverse process for data reconstruction using closed-form solutions and deep-learning- based solvers. However, these methodologies lack sufficient generalization and robustness, limiting their application to other seismic data domains. As a result, the entire DM must be retrained for each experiment, leading to high computational costs because of the model's complexity. We propose two diffusion-guided reconstruction solvers built on a fixed pretrained diffusion model. DM-DIGP combines reverse diffusion with a discriminator-regularized Deep Image Prior. In contrast, DM-RODIP combines reverse diffusion with a proximal operator conditioned by a Deep Image Prior that enforces measurement consistency. These approaches offer an alternative DM-based framework for seismic data reconstruction, in which the reverse process of a pretrained DM is guided to solve the reconstruction problem. We evaluated the proposed methods on both synthetic and field data, demonstrating superior reconstruction performance in PSNR and SSIM and outperforming state-of-the-art approaches.
- 30 September 2026
- Published by Society of Exploration Geophysicists in The Leading Edge
Guided Diffusion Models for Seismic Data Reconstruction via Deep Image Priors and Restoration Operators
Seismic data are widely used in subsurface energy exploration. However, several factors affect seismic data quality. Seismic data reconstruction is essential when acquisition is incomplete because missing sources or receivers produce subsampled shot-gathers with incomplete trace coverage that degrades data quality and affects subsequent seismic processing. Various deep-learning methodologies, particularly generative models, have been developed to reconstruct seismic signals. In this context, diffusion models (DMs) have emerged for seismic data processing, either by guiding image generation or by guiding the reverse process for data reconstruction using closed-form solutions and deep-learning- based solvers. However, these methodologies lack sufficient generalization and robustness, limiting their application to other seismic data domains. As a result, the entire DM must be retrained for each experiment, leading to high computational costs because of the model's complexity. We propose two diffusion-guided reconstruction solvers built on a fixed pretrained diffusion model. DM-DIGP combines reverse diffusion with a discriminator-regularized Deep Image Prior. In contrast, DM-RODIP combines reverse diffusion with a proximal operator conditioned by a Deep Image Prior that enforces measurement consistency. These approaches offer an alternative DM-based framework for seismic data reconstruction, in which the reverse process of a pretrained DM is guided to solve the reconstruction problem. We evaluated the proposed methods on both synthetic and field data, demonstrating superior reconstruction performance in PSNR and SSIM and outperforming state-of-the-art approaches.
- Open Access
- 19 May 2026
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 147
Change detection in hyperspectral and radar/multispectral images using an unmixing-based multivariate manifold estimation
- 19 May 2026
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 147
Open AccessChange detection in hyperspectral and radar/multispectral images using an unmixing-based multivariate manifold estimation
- Open Access
- 31 December 2025
- Published by Universidad Distrital Francisco Jose de Caldas in Revista Tecnura
- Vol. 29 (86) , 3-16
Aboveground Biomass Estimation of Colombian Cocoa Agroforestry Systems using Multispectral Images and Regression Methods
Context: Aboveground biomass (AGB) estimation for agricultural and environmental applications traditionally relies on time- consuming manual methods at ground level. Emerging approaches use remote sensing data evaluating spectral responses and vegetation indices in order to estimate the AGB more efficiently. Nonetheless, such techniques face various challenges in accounting for complex spatial distributions, especially when dealing with agroforestry systems (AFS). Objective: This work aimed to estimate AGB in a Colombian cocoa AFS from multispectral images acquired with an un-manned aerial vehicle (UAV). Methodology: In this work, the AGB was estimated by computing different vegetation indices from the measured spectral reflectances and evaluating two linear regression models, i.e., principal component regression (PCR) and partial least squares regression (PLSR), as well as a nonlinear regression model, a neural network composed by a perceptron with a single layer. Control points were obtained via on-ground biomass manual acquisition.Results: Numerical experiments resulted in a coefficient of determination of R2 = 0.58 for the best linear model, while the nonlinear model reached R2 = 0.86. Conclusions: AGB estimation in a cocoa AFS can be effectively performed using multispectral data acquired with a UAV and a simple nonlinear regression model.
- 31 December 2025
- Published by Universidad Distrital Francisco Jose de Caldas in Revista Tecnura
- Vol. 29 (86) , 3-16
Open AccessAboveground Biomass Estimation of Colombian Cocoa Agroforestry Systems using Multispectral Images and Regression Methods
Context: Aboveground biomass (AGB) estimation for agricultural and environmental applications traditionally relies on time- consuming manual methods at ground level. Emerging approaches use remote sensing data evaluating spectral responses and vegetation indices in order to estimate the AGB more efficiently. Nonetheless, such techniques face various challenges in accounting for complex spatial distributions, especially when dealing with agroforestry systems (AFS). Objective: This work aimed to estimate AGB in a Colombian cocoa AFS from multispectral images acquired with an un-manned aerial vehicle (UAV). Methodology: In this work, the AGB was estimated by computing different vegetation indices from the measured spectral reflectances and evaluating two linear regression models, i.e., principal component regression (PCR) and partial least squares regression (PLSR), as well as a nonlinear regression model, a neural network composed by a perceptron with a single layer. Control points were obtained via on-ground biomass manual acquisition.Results: Numerical experiments resulted in a coefficient of determination of R2 = 0.58 for the best linear model, while the nonlinear model reached R2 = 0.86. Conclusions: AGB estimation in a cocoa AFS can be effectively performed using multispectral data acquired with a UAV and a simple nonlinear regression model.
- Open Access
- 12 September 2023
- Published by Universidad Industrial de Santander in Revista UIS Ingenierías
- Vol. 22 (3) , 177-188
Shot-gather Reconstruction using a Deep Data Prior-based Neural Network Approach
Los levantamientos sísmicos usualmente se ven afectados por obstáculos o restricciones ambientales que impiden el muestreo regular en la adquisición sísmica. Por lo tanto, se han desarrollado diversos métodos para reconstruir estos datos faltantes, incluidos los métodos de aprendizaje profundo, los cuales permiten extraer características de información compleja, con la limitante de bases de datos sísmicos externos. Aunque otros trabajos se han enfocado principalmente en la reconstrucción de trazas, los disparos que no se pueden adquirir impactan directamente el flujo del procesamiento sísmico y representa un reto mayor en la regularización de datos sísmicos. En este trabajo proponemos DIPsgr, un método de reconstrucción de disparos sísmicos que usa solamente las medidas de las adquisiciones sísmicas incompletas para estimar la información faltante usando aprendizaje profundo no supervisado. Los experimentos numéricos con tres bases de datos muestran que DIPsgr recupera el conjunto completo de trazas en cada shot-gather, donde la información y los eventos sísmicos se conservan correctamente. Seismic surveys are often affected by environmental obstacles or restrictions that prevent regular sampling in seismic acquisition. To address missing data, various methods, including deep learning techniques, have been developed to extract features from complex information, albeit with the limitation of requiring external seismic databases. While previous works have primarily focused on trace reconstruction, missing shot-gathers directly impact the seismic processing flow and represent a major challenge in seismic data regularization. In this paper, we propose DIPsgr, a seismic shot-gather reconstruction method that uses only the incomplete seismic acquisition measurements to estimate their missing information employing unsupervised deep learning. Numerical experiments on three databases demonstrate that DIPsgr recovers the complete set of traces in each shot-gather, with preserved information and seismic events.
- 12 September 2023
- Published by Universidad Industrial de Santander in Revista UIS Ingenierías
- Vol. 22 (3) , 177-188
Open AccessShot-gather Reconstruction using a Deep Data Prior-based Neural Network Approach
Los levantamientos sísmicos usualmente se ven afectados por obstáculos o restricciones ambientales que impiden el muestreo regular en la adquisición sísmica. Por lo tanto, se han desarrollado diversos métodos para reconstruir estos datos faltantes, incluidos los métodos de aprendizaje profundo, los cuales permiten extraer características de información compleja, con la limitante de bases de datos sísmicos externos. Aunque otros trabajos se han enfocado principalmente en la reconstrucción de trazas, los disparos que no se pueden adquirir impactan directamente el flujo del procesamiento sísmico y representa un reto mayor en la regularización de datos sísmicos. En este trabajo proponemos DIPsgr, un método de reconstrucción de disparos sísmicos que usa solamente las medidas de las adquisiciones sísmicas incompletas para estimar la información faltante usando aprendizaje profundo no supervisado. Los experimentos numéricos con tres bases de datos muestran que DIPsgr recupera el conjunto completo de trazas en cada shot-gather, donde la información y los eventos sísmicos se conservan correctamente. Seismic surveys are often affected by environmental obstacles or restrictions that prevent regular sampling in seismic acquisition. To address missing data, various methods, including deep learning techniques, have been developed to extract features from complex information, albeit with the limitation of requiring external seismic databases. While previous works have primarily focused on trace reconstruction, missing shot-gathers directly impact the seismic processing flow and represent a major challenge in seismic data regularization. In this paper, we propose DIPsgr, a seismic shot-gather reconstruction method that uses only the incomplete seismic acquisition measurements to estimate their missing information employing unsupervised deep learning. Numerical experiments on three databases demonstrate that DIPsgr recovers the complete set of traces in each shot-gather, with preserved information and seismic events.
- 1 August 2023
- Published by Springer Nature
- p. 593-606
Encoded Diffractive Optics for Hyperspectral Imaging
- 1 August 2023
- Published by Springer Nature
- p. 593-606
Encoded Diffractive Optics for Hyperspectral Imaging
- 30 March 2022
- Published by Springer Nature
- p. 44-57
End-to-End Compressive Spectral Classification: A Deep Learning Approach Applied to the Grading of Tahiti Lime
- 30 March 2022
- Published by Springer Nature
- p. 44-57
End-to-End Compressive Spectral Classification: A Deep Learning Approach Applied to the Grading of Tahiti Lime
- 10 March 2020
- Published by Optica Publishing Group in Optics Express
- Vol. 28 (6) , 8528-8540
Coupled deep learning coded aperture design for compressive image classification
A coupled deep learning approach for coded aperture design and single-pixel measurements classification is proposed. A whole neural network is trained to simultaneously optimize the binary sensing matrix of a single-pixel camera (SPC) and the parameters of a classification network, considering the constraints imposed by the compressive architecture. Then, new single-pixel measurements can be acquired and classified with the learned parameters. This method avoids the reconstruction process while maintaining classification reliability. In particular, two network architectures were proposed, one learns re-projected measurements to the image size, and the other extracts small features directly from the compressive measurements. They were simulated using two image data sets and a test-bed implementation. The first network beats in around 10% the accuracy reached by the state-of-the-art methods. A 2x increase in computing time is achieved with the second proposed net.
- 10 March 2020
- Published by Optica Publishing Group in Optics Express
- Vol. 28 (6) , 8528-8540
Coupled deep learning coded aperture design for compressive image classification
A coupled deep learning approach for coded aperture design and single-pixel measurements classification is proposed. A whole neural network is trained to simultaneously optimize the binary sensing matrix of a single-pixel camera (SPC) and the parameters of a classification network, considering the constraints imposed by the compressive architecture. Then, new single-pixel measurements can be acquired and classified with the learned parameters. This method avoids the reconstruction process while maintaining classification reliability. In particular, two network architectures were proposed, one learns re-projected measurements to the image size, and the other extracts small features directly from the compressive measurements. They were simulated using two image data sets and a test-bed implementation. The first network beats in around 10% the accuracy reached by the state-of-the-art methods. A 2x increase in computing time is achieved with the second proposed net.
- Open Access
- 26 January 2020
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 32 (14) , 59-1
Single-shot Coded Diffraction System for 3D Object Shape Estimation
Fast track article for IS&T International Symposium on Electronic Imaging 2020: Computational Imaging proceedings.
- 26 January 2020
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 32 (14) , 59-1
Open AccessSingle-shot Coded Diffraction System for 3D Object Shape Estimation
Fast track article for IS&T International Symposium on Electronic Imaging 2020: Computational Imaging proceedings.
- Open Access
- 22 February 2019
- Published by Optica Publishing Group in Applied Optics
- Vol. 58 (7) , B28-B38
Shifting colored coded aperture design for spectral imaging
Compressive spectral imaging (CSI) systems sense 3D spatio-spectral data cubes with just a few two-dimensional (2D) projections by using a coded aperture, a dispersive element, and a focal plane array (FPA). The coded apertures in these systems, whose main function is the modulation of the data cube, are often implemented through photomasks attached to piezoelectric devices. A remarkable improvement on this configuration has been recently proposed, the replacement of the block-unblock coded apertures by patterned optical filter arrays, referred to as “colored” coded apertures, which allow spatial and spectral modulation. When using these colored coded apertures, its real implementation in terms of cost and complexity directly depends on the number of filters to be used, as well as the number of shots to be captured. A shifting colored coded aperture optimization featuring these observations is proposed, with the aim to improve the imaging quality reconstruction and to generate an achievable optical implementation with a limited number of filters requiring only one mask to acquire any number of shots. The mathematical model of the computational imaging strategy to overcome the practical limitations of actual CSI systems is presented along with a testbed implementation. Simulations, as well as experimental results, will prove the accuracy and performance of the proposed shifting colored coded aperture design over the current literature designs.
- 22 February 2019
- Published by Optica Publishing Group in Applied Optics
- Vol. 58 (7) , B28-B38
Open AccessShifting colored coded aperture design for spectral imaging
Compressive spectral imaging (CSI) systems sense 3D spatio-spectral data cubes with just a few two-dimensional (2D) projections by using a coded aperture, a dispersive element, and a focal plane array (FPA). The coded apertures in these systems, whose main function is the modulation of the data cube, are often implemented through photomasks attached to piezoelectric devices. A remarkable improvement on this configuration has been recently proposed, the replacement of the block-unblock coded apertures by patterned optical filter arrays, referred to as “colored” coded apertures, which allow spatial and spectral modulation. When using these colored coded apertures, its real implementation in terms of cost and complexity directly depends on the number of filters to be used, as well as the number of shots to be captured. A shifting colored coded aperture optimization featuring these observations is proposed, with the aim to improve the imaging quality reconstruction and to generate an achievable optical implementation with a limited number of filters requiring only one mask to acquire any number of shots. The mathematical model of the computational imaging strategy to overcome the practical limitations of actual CSI systems is presented along with a testbed implementation. Simulations, as well as experimental results, will prove the accuracy and performance of the proposed shifting colored coded aperture design over the current literature designs.
- Open Access
- 5 January 2018
- Published by Universidad Industrial de Santander in Revista UIS Ingenierías
- Vol. 17 (1) , 209-216
Multiresolution-based reconstruction for compressive spectral video sensing using a spectral multiplexing sensor
- 5 January 2018
- Published by Universidad Industrial de Santander in Revista UIS Ingenierías
- Vol. 17 (1) , 209-216
Open AccessMultiresolution-based reconstruction for compressive spectral video sensing using a spectral multiplexing sensor