U-Net based neural network for fringe pattern denoising

Gurrola-Ramos, Javier; Dalmau, Oscar; Alarcón, Teresa E.

doi:10.1016/j.optlaseng.2021.106829

Cited by 19 publications

(8 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous studies have confirmed that U-Net-based models are excellent tools for image reconstruction and denoising (Reymann et al 2019, Gurrola-Ramos et al 2022. In particular, FUS-Net, which uses 2D convolutional computation, is more accurate in restoring information obscured by noise and offers greater flexibility than methods based on notch filtering (Lee and Konofagou 2021).…”

Section: Discussionmentioning

confidence: 91%

Suppressing HIFU interference in ultrasound images using 1D U-Net-based neural networks

Yang,

Li,

Liu

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective. One big challenge with high-intensity focused ultrasound (HIFU) is that the intense acoustic interference generated by HIFU irradiation overwhelms the B-mode monitoring images, compromising monitoring effectiveness. This study aims to overcome this problem using a one-dimensional (1D) deep convolutional neural network. Approach. U-Net-based networks have been proven to be effective in image reconstruction and denoising, and the two-dimensional (2D) U-Net has already been investigated for suppressing HIFU interference in ultrasound monitoring images. In this study, we propose that the one-dimensional (1D) convolution in U-Net-based networks is more suitable for removing HIFU artifacts and can better recover the contaminated B-mode images compared to 2D convolution. Ex-vivo and in-vivo HIFU experiments were performed on a clinically equivalent ultrasound-guided HIFU platform to collect image data, and the 1D convolution in U-Net, Attention U-Net, U-Net++, and FUS-Net was applied to verify our proposal. Main results. All 1D U-Net-based networks were more effective in suppressing HIFU interference than their 2D counterparts, with over 30% improvement in terms of structural similarity (SSIM) to the uncontaminated B-mode images. Additionally, 1D U-Nets trained using ex-vivo datasets demonstrated better generalization performance in in-vivo experiments. Significance. These findings indicate that the utilization of 1D convolution in U-Net-based networks offers great potential in addressing the challenges of monitoring in ultrasound-guided HIFU systems.

show abstract

Section: Discussionmentioning

confidence: 91%

Suppressing HIFU interference in ultrasound images using 1D U-Net-based neural networks

Yang,

Li,

Liu

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

“…: 1200 pairs --- Reyes-Figueroa et al 86 Noisy fringe pattern Noise-free fringe pattern U-Net and ResNet Sim. : 25,000 pairs l 1 -norm Gurrola-Ramos et al 87 Noisy fringe pattern Noise-free fringe pattern U-Net and DenseNet Sim. : 1500 pairs l 1 -norm Hologram generation Zhang et al 88 , 89 Hologram Phase-shifting holograms Y-Net Sim.…”

Section: Dl-pre-processing For Phase Recoverymentioning

confidence: 99%

“…Reyes-Figueroa et al 86 further showed that the U-Net and its improved version (V-Net) are better than DnCNN for fringe pattern denoising, because their proposed V-Net has more channels on the outer side than on the inner side, retaining more details. Given the U-Net’s outstanding mapping capabilities, Gurrola-Ramos et al 87 also improved it for fringe pattern denoising, where dense blocks are leveraged for reusing feature layers, local residual learning is used to address the vanishing gradient problem, and global residual learning is used to estimate the noise of the image instead of the denoised image directly. Compared with other neural networks mentioned above, it has a minor model complexity while maintaining the highest accuracy.…”

Section: Dl-pre-processing For Phase Recoverymentioning

confidence: 99%

On the use of deep learning for phase recovery

Wang,

Song,

Wang

et al. 2024

Light Sci Appl

View full text Add to dashboard Cite

Phase recovery (PR) refers to calculating the phase of the light field from its intensity measurements. As exemplified from quantitative phase imaging and coherent diffraction imaging to adaptive optics, PR is essential for reconstructing the refractive index distribution or topography of an object and correcting the aberration of an imaging system. In recent years, deep learning (DL), often implemented through deep neural networks, has provided unprecedented support for computational imaging, leading to more efficient solutions for various PR problems. In this review, we first briefly introduce conventional methods for PR. Then, we review how DL provides support for PR from the following three stages, namely, pre-processing, in-processing, and post-processing. We also review how DL is used in phase image processing. Finally, we summarize the work in DL for PR and provide an outlook on how to better use DL to improve the reliability and efficiency of PR. Furthermore, we present a live-updating resource (https://github.com/kqwang/phase-recovery) for readers to learn more about PR.

show abstract

“…Currently, the Unet deep learning network has been extensively applied for optical metrology for various tasks, from pattern analysis [19], phase retrieval [20], phase unwrapping [21], and surface reconstruction [15]. Unet was originally developed for image segmentation tasks for biomedical images [22].…”

Section: Introductionmentioning

confidence: 99%

DYnet++: A Deep Learning Based Single-Shot Phase-Measuring Deflectometry for the 3-D Measurement of Complex Free-Form Surfaces

Nguyen

Ghim

Rhee

2024

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

Deflectometry is a three-dimensional optical technique based on structured light projection for measuring specular free-form surfaces. Much research on single-shot deflectometry has been conducted to develop reliable measurement techniques, even for harsh industrial environments. But the measurement of complex surfaces with low reflectivity is still a big challenge in single-shot deflectometry since it is very tricky to obtain the phase from a poor quality single complex pattern. Inspired by recent applications of deep learning in the optical metrology field, we propose a novel single-shot deflectometry that utilizes deep learning to measure complex surfaces with low quality in just a single shot. A deformable mirror with 9 actuators was developed to generate big data for various surface shapes to train the deep learning model. We developed a deep learning network model, DYnet++, to retrieve the phase from single composite patterns even where both closed-and openedloops are presented. We verified the feasibility of the proposed deep learning based single-shot deflectometry by comparing measurement results with those from 16 step phase-shifting method.

show abstract

U-Net based neural network for fringe pattern denoising

Cited by 19 publications

References 24 publications

Suppressing HIFU interference in ultrasound images using 1D U-Net-based neural networks

Suppressing HIFU interference in ultrasound images using 1D U-Net-based neural networks

On the use of deep learning for phase recovery

DYnet++: A Deep Learning Based Single-Shot Phase-Measuring Deflectometry for the 3-D Measurement of Complex Free-Form Surfaces

Contact Info

Product

Resources

About