A review and meta-analysis of Generative Adversarial Networks and their applications in remote sensing

Jozdani, Shahab; Chen, Dongmei; Pouliot, Darren; Johnson, Brian Alan

doi:10.1016/j.jag.2022.102734

Cited by 31 publications

(20 citation statements)

References 173 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They have also reviewed the datasets along with evaluation metrics and highlighted the issues faced by different speech GANs. Jozdani et al [51] have presented a systematic review along with meta-analysis of GANbased studies in remote sensing. Authors have also evaluated the GAN theories, applications, and difficulties and identified the research gaps that need to be addressed in the future by remote sensing researchers.…”

Section: Related Workmentioning

confidence: 99%

Mathematical analysis of loss function of GAN and its loss function variants

2022

IJATEE

View full text Add to dashboard Cite

Generative adversarial networks (GANs) have turned up as the most widely used approaches for creating realistic samples. They're the effective latent variable models for learning complex real distributions. However, despite their enormous success and popularity, the process of training GANs remains challenging and suffers from a number of failures. These failures include mode collapse where the generator generates the same set of output for different inputs which finally leads to loss of diversity; non-convergence because of the diverging and oscillatory behaviors of both generator and discriminator while being trained; and vanishing or exploding gradients due to which either no learning or extremely slow learning takes place. In the past years, a variety of strategies for stabilizing GAN training have been explored which includes modified architectures, loss functions, and other methods. The choice of loss function has been found to be the most crucial part of the GAN model because it influences the vanishing gradient and model collapse directly. Viewing these loss functions as divergence minimization has provided a rich avenue of development. All of these factors make GAN training inherently unstable, and this instability is difficult to comprehend mathematically. This paper intends to provide a thorough mathematical explanation of these divergence minimization functions. It illustrates in great detail the two variants of the loss functions of the original GAN, their optimization to kullback-leibler (KL) divergence and jensen-shannon (JS) divergence along with their shortcomings. It also describes the loss functions of the different loss function GAN variants that have been proposed to mitigate these shortcomings as well as their minimization. The original GAN and its loss function variants have also been implemented using the standard MNIST, Fashion-MNIST, and CIFAR-10 datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Mathematical analysis of loss function of GAN and its loss function variants

2022

IJATEE

View full text Add to dashboard Cite

show abstract

“…GANs are undoubtedly one of the most creative advances in deep learning (DL) in recent years (Goodfellow et al, 2014;Jozdani et al, 2022). GANs are based on the min-max, zero-sum game theory.…”

Section: Generative Adversarial Network (Gan)mentioning

confidence: 99%

“…Fortunately, the RS community quickly recognized the value of GANs and successfully adopted them in RS image reconstruction/restoration, RS image denoising, RS data translation and other RS-related tasks. Jozdani et al (Jozdani et al, 2022) reviewed the relevant research on GANs in the context of RS, expecting to help the RS community understand the potential and limitations of GANs in this field. Although there have been several studies on GANs in the context of RS in recent years, remote sensing research on the applications of GANs in the ocean is still insufficient.…”

Section: Generative Adversarial Network (Gan)mentioning

confidence: 99%

“…There are 5478 sets of sea surface parameter combinations (SST, SSTA, SSHA) after removing some daily combinations with poor quality grid points for one or more parameters. For a GAN-based model, it is difficult to measure the quality of generated data in an objective manner while training, as the loss value of GANs may not be indicative of the model's performance during training(Jozdani et al, 2022). The maximum number of training epochs is set to 200, and the model with the best performance on the validation dataset is saved as the final model.Second, the trained model at each depth level was fine-tuned with satellite remote sensing data and CORA 2.0 data.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Deriving Sea Subsurface Temperature Fields from Satellite Remote Sensing Data Using a Generative Adversarial Network Model

Zhang

Ning

Zhang

et al. 2022

Preprint

View full text Add to dashboard Cite

Ingenious use of multisource satellite observations to accurately invert global and regional subsurface thermohaline structure is essential for understanding ocean interior processes, but extremely challenging. This study proposes a new method from the sea surface information inverting daily subsurface temperature (ST) based on generative adversarial network (GAN) model in China's marginal seas. The proposed GAN-based model can project the STs from sea surface information (SLA, SSTA, SST) with a high resolution of 1/12°. A traditional regression-based model, Modular Ocean Data Assimilation System (MODAS), is set up the same experiments for the sake of comparison. The results show that the averaged RMSE results are less than 1.45°C in upper 200m and the highest averaged R2 of 0.97 at the 70m level, which are better than that of MODAS. Errors analysis and typical oceanographic phenomena analysis results show the superiority of the proposed GAN-based model in this study. This study can provide high-precision daily ST data from sea surface information, which can be expanded to further studies on the ocean interior variation characteristics.

show abstract

“…Recent studies that have focused on spectral classification methods include neural networks (NN) [5], generative adversarial networks (GANs) for remote sensing [6], apart from the classical support vector machine (SVM) [7] and random forest (RF) [8] technique. Although the supervised learning algorithms for image classification show promising results [9], these methods are heavily reliant on large high-quality labeled training data sets.…”

Section: Introductionmentioning

confidence: 99%

Tree species classification from hyperspectral data using graph-regularized neural networks

Bandyopadhyay¹,

Mukherjee²

2022

Preprint

View full text Add to dashboard Cite

Manual labeling of tree species remains a challenging task, especially in tropical regions, owing to inaccessibility and labor-intensive ground-based surveys. Hyperspectral images (HSIs), through their narrow and contiguous bands, can assist in distinguishing tree species based on their spectral properties. Therefore, automated classification algorithms on HSI images can help augment the limited labeled information and generate a real-time classification map for various tree species. Achieving high classification accuracy with a limited amount of labeled information in an image is one of the key challenges that researchers have started addressing in recent years. We propose a novel graph-regularized neural network (GRNN) algorithm that encompasses the superpixel-based segmentation for graph construction, a pixel-wise neural network classifier, and the label propagation technique to generate an accurate classification map. GRNN outperforms several state-of-the-art techniques not only for the standard Indian Pines HSI but also achieves a high classification accuracy (approx. 92%) on a new HSI data set collected over the forests of French Guiana (FG) even when less than 1% of the pixels are labeled. We show that GRNN is not only competitive with the state-of-the-art semi-supervised methods, but also exhibits lower variance in accuracy for different number of training samples and over different independent random sampling of the labeled pixels for training.

show abstract

A review and meta-analysis of Generative Adversarial Networks and their applications in remote sensing

Cited by 31 publications

References 173 publications

Mathematical analysis of loss function of GAN and its loss function variants

Mathematical analysis of loss function of GAN and its loss function variants

Deriving Sea Subsurface Temperature Fields from Satellite Remote Sensing Data Using a Generative Adversarial Network Model

Tree species classification from hyperspectral data using graph-regularized neural networks

Contact Info

Product

Resources

About