Fast generative adversarial networks model for masked image restoration

Cao, Zhiyi; Niu, Shaozhang; Zhang, Jiwei; Wang, Xinyi

doi:10.1049/iet-ipr.2018.5592

Cited by 9 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the convenience of calculation, we assume the mathematical difference q (between visible watermarked image and real image) which is from 0.01 to 1. Inspired by [11], we can get the pixel difference calculated by the GANs model:

y^{GANs} = log ()(, q) \times 0.1 - log ()(, 1 - q) \times 0.1 + 1 .

To the same, we will get the pixel difference concerning the DCGAN model:

y^{DCGAN} = y^{GANs} .

Inspired by [11], we can get the pixel difference calculated by the WGAN model:

y^{WGAN} = q .

To the same, we will get the pixel difference concerning the pix2pix model:

y^{pix 2 pix} = y^{WGAN} .

Inspired by [11], we can get the pixel difference calculated by the LSGAN model:

y^{LSGAN} = q^{2} .

According to (4), the pixel difference calculated by the VWGAN model can be defined as follows:

y^{VWGAN} = 2 q - q^{2}

The visible watermarked images and the real images are very similar in most condition. In other words, the mathematical differences are minimal in most condition.…”

Section: Proposed Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Generative adversarial networks model for visible watermark removal

et al. 2019

Self Cite

View full text Add to dashboard Cite

Previously visible watermark removal algorithms required the location of known watermarks. A corresponding removal algorithm is then proposed based on the location and the feature of the watermark. If the location of the watermark is random or the watermark has different angles, the watermark removal algorithm will encounter problems. The authors recommend a visible watermark removal algorithm based on generative adversarial networks (GANs) and self‐attention mechanisms. During the training, the authors introduce a GANs model to build mappings between watermarked images and real images. The authors observe that the feature of the watermarked region in different watermarked images is invariant in nature, and the other regions are changed. The self‐attention layer will automatically focus on this invariant feature. Experiments on two public datasets prove that the authors’ model has gained excellent performance. Compared with the other four most competitive watermark removal models, the authors improve the watermark removal rate indicator from 17 to 92%. For the other four evaluation indicators, the authors have improved performance by up to 20%.

show abstract

y^{GANs} = log ()(, q) \times 0.1 - log ()(, 1 - q) \times 0.1 + 1 .

To the same, we will get the pixel difference concerning the DCGAN model:

y^{DCGAN} = y^{GANs} .

Inspired by [11], we can get the pixel difference calculated by the WGAN model:

y^{WGAN} = q .

To the same, we will get the pixel difference concerning the pix2pix model:

y^{pix 2 pix} = y^{WGAN} .

Inspired by [11], we can get the pixel difference calculated by the LSGAN model:

y^{LSGAN} = q^{2} .

According to (4), the pixel difference calculated by the VWGAN model can be defined as follows:

y^{VWGAN} = 2 q - q^{2}

The visible watermarked images and the real images are very similar in most condition. In other words, the mathematical differences are minimal in most condition.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…As they are unable to locate the watermarks they cannot be used for watermark removal. Recently, a novel fast GANs model [11] for complex masked image restoration was proposed. Its neighbouring network requires the location of the watermark to remove the watermark to the greatest extent.…”

Section: Related Workmentioning

confidence: 99%

Generative adversarial networks model for visible watermark removal

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, we introduce two parsing networks for our MRGAN model to find more parsing differences between the generated image and the real image. The model most similar to our model is the fast GANs model [23]. Although the fast GANs model proposes a novel neighbouring module compared with our model, it focuses on the restoration of local masked images.…”

Section: Related Workmentioning

confidence: 99%

“…Different from the FACEGAN model, the parsing networks are used to compare the similarity of two images, not to enhance image quality by using the pre‐trained VGG19 [26] model. Different from the fast GANs model [23], for global mosaic removal, we compare image difference in three convolutional layers and all the convolutional layers. Inspired by [27, 28], features are extracted on convolutional layers conv3‐4, conv4‐4, and conv5‐4.…”

Section: Mrgan Model Designmentioning

confidence: 99%

MRGAN: a generative adversarial networks model for global mosaic removal

et al. 2020

Self Cite

View full text Add to dashboard Cite

“…Generative adversarial networks (GAN) [20] is one of the most promising deep learning methods in complex distribution in recent years. It is widely used in the field of computer vision [21][22][23][24]. The shadow removal method based on GAN has achieved good results recently [25].…”

Section: Introductionmentioning

confidence: 99%

Shadow removal method of soil surface image based on GAN used for estimation of farmland soil moisture content

Meng

Yang

Wang

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

It is important to obtain soil moisture content (SMC) in farmland, and soil surface images can be used to rapidly estimate SMC. The objective of this study was to propose a shadow removal algorithm to eliminate the effect of shadows in soil surface images，so as to improve the accuracy of SMC estimation. The structure of the proposed Soil Shadow Generative Adversarial Networks (SS GAN) was a circulating network, which is an unsupervised method and does not require paired shadow image sets for network training. Four loss functions were defined for the network to effectively remove shadows and ensure texture detail and color consistency. This method is compared with traditional methods, supervised and unsupervised deep learning techniques by comparative experiments. Evaluations were made from visual and quantitative comparisons. Visually, the best shadow removal method was proved, it almost has no shadow boundaries or shadow areas visible for samples. The Peak Signal to Noise Ratio (PSNR) and Structural Similarity (SSIM) were used to quantitatively compare shadow removal images with real non-shadow images. The PSNR and SSIM of SS GAN were 28.46 and 0.95 respectively, which are superior to other methods, indicating that the images processed by SS GAN were closer to the real non-shadow images. Field experiments results shown that SS GAN has excellent shadow removal performance in the self-developed vehicle-mounted detection system. In order to verify the improvement effect of shadow removal image on SMC estimation accuracy, further field test was conducted to estimate SMC. Compared with SMC estimation results before and after shadow removal, R2 increased from 0.69 to 0.76, and Root Mean Square Error (RMSE)decreased from 1.39 to 0.94%. The results show that the proposed method can effectively remove the shadow of soil image and improve the accuracy of SMC estimation in farmland.

show abstract

Fast generative adversarial networks model for masked image restoration

Cited by 9 publications

References 18 publications

Generative adversarial networks model for visible watermark removal

Generative adversarial networks model for visible watermark removal

MRGAN: a generative adversarial networks model for global mosaic removal

Shadow removal method of soil surface image based on GAN used for estimation of farmland soil moisture content

Contact Info

Product

Resources

About