Wrapped phase denoising using convolutional neural networks

Yan, Kun; Yu, Yingjie; Sun, Tao; Asundi, Anand Krishna; Kemao, Qian

doi:10.1016/j.optlaseng.2019.105999

Cited by 61 publications

(26 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, the ability of the network to generate holograms will be more accurate and faster. A similar approach of training by intermediate variables instead of its original target data has been validated by Ketao Yan et al [31], using convolutional neural networks in wrapped bit-phase denoising. During the training process, the model can calculate holograms quickly within a predicable time window which only depends on the model size (number of convolutional layers, number of kernels, kernel size), the number of model iterations, the number of depth planes and the size of the hologram.…”

Section: Neural Network Algorithm and Its Training Strategymentioning

confidence: 99%

“…The plane 3 is farthest from the hologram plane at a distance of 100 mm (d 3 ) and all planes are spaced at 1 mm. The multi-plane iterative ASM is used to generate the complex amplitudes on the hologram plane according to the theory described in Section 2.1 [27,31], which is implemented by MATLAB. The hologram plane pixel spacing and the wavelength of the light source are set to 8 µm and 632.8 nm, respectively.…”

Section: Neural Network Algorithm and Its Training Strategymentioning

confidence: 99%

See 1 more Smart Citation

Computing 3D Phase-Type Holograms Based on Deep Learning Method

Zheng

Zhou

et al. 2021

Photonics

View full text Add to dashboard Cite

Computer holography is a technology that use a mathematical model of optical holography to generate digital holograms. It has wide and promising applications in various areas, especially holographic display. However, traditional computational algorithms for generation of phase-type holograms based on iterative optimization have a built-in tradeoff between the calculating speed and accuracy, which severely limits the performance of computational holograms in advanced applications. Recently, several deep learning based computational methods for generating holograms have gained more and more attention. In this paper, a convolutional neural network for generation of multi-plane holograms and its training strategy is proposed using a multi-plane iterative angular spectrum algorithm (ASM). The well-trained network indicates an excellent ability to generate phase-only holograms for multi-plane input images and to reconstruct correct images in the corresponding depth plane. Numerical simulations and optical reconstructions show that the accuracy of this method is almost the same with traditional iterative methods but the computational time decreases dramatically. The result images show a high quality through analysis of the image performance indicators, e.g., peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and contrast ratio. Finally, the effectiveness of the proposed method is verified through experimental investigations.

show abstract

Section: Neural Network Algorithm and Its Training Strategymentioning

confidence: 99%

Section: Neural Network Algorithm and Its Training Strategymentioning

confidence: 99%

Computing 3D Phase-Type Holograms Based on Deep Learning Method

Zheng

Zhou

et al. 2021

Photonics

View full text Add to dashboard Cite

show abstract

“…The most relevant aspect designing the applications of NN models to physical problems is the design of the training set. Using Zernike polynomials we followed the method designed in [6] to generate a dataset of 10.000 train images with size 128x128 pixels, based in randomly generate the coefficients for the first 14 Zernike polynomials. This set was the ground true for the inputs and for the noisy images in the training dataset, we included an additive and random noise component based on a normal distribution to align the noisy images to the typical images obtained in speckle pattern interference.…”

Section: Trainingmentioning

confidence: 99%

Fringe Pattern Denoising using U-Net based neural network

Crespo¹,

Moreno²,

Rabuñal³

et al. 2020

EPJ Web Conf.

View full text Add to dashboard Cite

show abstract

“…To the best of our knowledge the first attempt has been proposed in [10], where the PU problem was converted into a segmentation task. In Yan et al [11], the authors embedded phase denoising and wrap count reconstruction in a single framework. A similar approach for InSAR unwrapping has been proposed in [12] with the difference that instead of absolute wrap count values, its gradient is reconstructed.…”

Section: Introductionmentioning

confidence: 99%

A CNN-Based Coherence-Driven Approach for InSAR Phase Unwrapping

Sica

Cellini

Scarpa

et al. 2022

IEEE Geosci. Remote Sensing Lett.

View full text Add to dashboard Cite

Phase unwrapping (PU) is among the most critical tasks in synthetic aperture radar (SAR) interferometry (InSAR). Due to the presence of noise, the interferogram usually presents phase inconsistencies, also called residues, which imply a nonunivocal solution. This work investigates the PU problem from a semantic segmentation perspective by exploiting convolutional neural network (CNN) models. In particular, by exploiting a popular deep-learning architecture, we introduce the interferometric coherence as an input feature and analyze the performance increase against classical methods. For the network training, we generate a variegated data set by introducing a controlled number of phase residues, and considering both synthetic and real InSAR data. Eventually, we compare the proposed method to state-of-the-art algorithms on synthetic and real InSAR data taken from the TanDEM-X mission, obtaining encouraging results.

show abstract