PhaseNet 2.0: Phase Unwrapping of Noisy Data Based on Deep Learning Approach

Spoorthi, G. E.; Gorthi, Rama Krishna Sai Subrahmanyam; Gorthi, Subrahmanyam

doi:10.1109/tip.2020.2977213

Cited by 162 publications

(66 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In both occasions, the models were trained using ADAM optimizer with a learning rate of 0.001 and they converged within 10 epochs taking only ∼ 1.5 hours to train. Similarly, Ryu et al's [18] network, PhaseNet 2.0 [8] and QGPU [5] were implemented and tested on the two datasets as well. Out of these, Ryu et al's network and PhaseNet 2.0 were trained on both noisy and noise free datasets.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…In recent years, deep learning algorithms have gained popularity and achieved state-of-the-art performance in many computer vision tasks. Following this trend, a few recent studies [8,9,10,11] have attempted to apply deep learning to address the phase unwrapping problem. Out of these, [8,9,10] have reformulated the phase unwrapping problem as a semantic segmentation task where Fully Convolutional Networks (FCNs) are trained to predict the wrap count at each pixel.…”

Section: Introductionmentioning

confidence: 99%

“…Among these methods, Spoorthi et. al's [8] PhaseNet 2.0, a deep encoder-decoder architecture comprised of dense blocks [12], has the best phase unwrapping performance. To the best of our knowledge only Wang et al [11] has considered phase unwrapping as a regression problem where an FCN inspired by U-NET [13] and ResNet [14] is employed to directly estimate the true phase from wrapped phase.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Joint Convolutional and Spatial Quad-Directional LSTM Network for Phase Unwrapping

Perera

Silva

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Phase unwrapping is a classical ill-posed problem which aims to recover the true phase from wrapped phase. In this paper, we introduce a novel Convolutional Neural Network (CNN) that incorporates a Spatial Quad-Directional Long Short Term Memory (SQD-LSTM) for phase unwrapping, by formulating it as a regression problem. Incorporating SQD-LSTM can circumvent the typical CNNs' inherent difficulty of learning global spatial dependencies which are vital when recovering the true phase. Furthermore, we employ a problem specific composite loss function to train this network. The proposed network is found to be performing better than the existing methods under severe noise conditions (Normalized Root Mean Square Error of 1.3% at SNR = 0 dB) while spending a significantly less computational time (0.054s). The network also does not require a large scale dataset during training, thus making it ideal for applications with limited data that require fast and accurate phase unwrapping.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Joint Convolutional and Spatial Quad-Directional LSTM Network for Phase Unwrapping

Perera

Silva

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…According to existing research [11][12][13][14][15][16][17][18], the key point of DLbased phase unwrapping is to determine an appropriate learning object (i.e., the ground truth), so as to investigate the PU problem from deep learning perspective. Through clever transformation, phase unwrapping can be converted into an image segmentation task, which is DL's area of expertise.…”

Section: Deep Learning Based Algorithmsmentioning

confidence: 99%

“…Wang et al [15] used the absolute phase instead of ambiguity number K as ground truth and proposed a DLPU structure, which can directly extract the absolute phase from the wrapped phase after training without any post-processing step. Spoorthi et al [16] proposed PhaseNet 2.0 on the basis of PhaseNet, utilized DenseNet to implement phase unwrapping tasks, which greatly increases the classes of ambiguity number that can be accurately predicted and enhances the network's performance by proposing a new Residual Loss.…”

Section: Introductionmentioning

confidence: 99%

A Novel Quality-Guided Two-Dimensional InSAR Phase Unwrapping Method via GAUNet

Wang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Phase unwrapping (PU) has always been a critical and challenging step in interferometric synthetic aperture radar (InSAR) data processing. Inspired by existing research, i.e., the PGNet, we propose a novel quality-guided 2-D InSAR PU method via deep learning, and regard PU as a two-stage process. In the first stage, the ambiguity gradient is estimated using the proposed global attention U-Net (GAUNet) architecture, which combines the classic U-Net structure and global attention mechanism. Then in the second stage, the classical PU framework (e.g., the L1-or L2-norm) is applied as a post-processing operation to retrieve the absolute phase. Since class imbalance is a key factor affecting the estimation of ambiguity gradient, different strategies based on four commonly used quality maps are adopted to deal with the problem. The quality map is not only input as additional information for the guidance of the training process, but also participates in the construction of loss function. As a result, GAUNet can pay more attention to the non-zero ambiguity gradients. By using the number of residues as the evaluation metric, we can choose the optimum strategy for the restoration of the absolute phase. In addition to the simulated interferograms, the proposed method is tested both on a real topographic interferogram exhibiting rugged topography and phase aliasing and a differential interferogram measuring the deformation from MW 6.9 Hawaii earthquake, all yield state-of-art performance when comparing with the widely used traditional 2-D PU methods.

show abstract

Automated Phase Unwrapping in DHM with Deep Learning

2022

Artificial Intelligence in Digital Holographic Imaging

View full text Add to dashboard Cite

PhaseNet 2.0: Phase Unwrapping of Noisy Data Based on Deep Learning Approach

Cited by 162 publications

References 44 publications

A Joint Convolutional and Spatial Quad-Directional LSTM Network for Phase Unwrapping

A Joint Convolutional and Spatial Quad-Directional LSTM Network for Phase Unwrapping

A Novel Quality-Guided Two-Dimensional InSAR Phase Unwrapping Method via GAUNet

Automated Phase Unwrapping in DHM with Deep Learning

Contact Info

Product

Resources

About