Attention-guided multi-path cross-CNN for underwater image super-resolution

Zhang, Yan; Yang, Shangxue; Sun, Yu; Liu, Shudong; Li, Xianguo

doi:10.1007/s11760-021-01969-4

Cited by 16 publications

(18 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AHTCN achieves the best PSNR and SSIM scores at scale factors of 2×, 4×, and 8× and ranks among the top two in terms of UIQM. At a scale factor of 4×, our method achieves a 0.56 dB improvement in PSNR and a 0.05 improvement in SSIM compared with AMPCNet 17 . In addition, AHTCN has 44.4% fewer model parameters.…”

Section: Resultsmentioning

confidence: 94%

“…The comparison includes several SOTA methods, such as SRCNN, 11 DSRCNN, 35 EDSRGAN, 13 SRResNet, 14 SRGAN, 14 ESRGAN, 36 SRDRM, 10 SRDRM-GAN, 10 AMPCNet, 17 and RDLN 37 . Moreover, the results are evaluated using three metrics: peak signal-to-noise ratio (PSNR), 38 structure similarity index (SSIM), 39 and underwater image quality measure (UIQM) 40 .…”

Section: Resultsmentioning

confidence: 99%

“…At a scale factor of 4×, our method achieves a 0.56 dB improvement in PSNR and a 0.05 improvement in SSIM compared with AMPCNet. 17 In addition, AHTCN has 44.4% fewer model parameters. The adoption of a two-branch structure with a hybrid CNN and transformer in AHTCN is proven to achieve superior results while also reducing the number of parameters and decreasing the model size.…”

Section: Quantitative Evaluationmentioning

confidence: 99%

“…Then Cherian and Poovammal 16 proposed AlphaSRGAN, which introduces a preprocessing method and optimizes the network structure to improve the quality of reconstructed images. Zhang et al 17 proposed an attention-guided multipath cross-convolutional neural network (AMPCNet) for underwater images, which makes full use of both attention and internal information to enhance the feature learning ability of the model. Moreover, Wang et al 18 proposed a novel progressive frequency-interleaved network to exploit the frequency domain features of images.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Attention-guided hybrid transformer-convolutional neural network for underwater image super-resolution

Zhan,

Li,

Zhang

2024

J. Electron. Imag.

View full text Add to dashboard Cite

Underwater images suffer from localized distortion and blurred degradation of edge structures due to light absorption and scattering by water. However, existing superresolution (SR) methods for underwater images cannot effectively solve the above problems and encounter model sizes that are too large. To this end, we propose an attention-guided hybrid transformer-CNN network (AHTCN) to improve the SR reconstruction of underwater images through the interaction of local and multiscale global information, as well as the long-range dependencies modeling capability. Specifically, AHTCN mainly consists of several cascaded transformer-CNN feature extraction blocks (TCFEB) and an image reconstruction module. In TCFEB, the designed attention-based channel separation mechanism can adaptively separate the weighted features while reducing the number of model parameters and then extract the local details and global structural information at different scales through the dual-stream structure. Moreover, we replace the feedforward layer in the transformer with the blueprint separable convolutional feedforward layer and propose an enhanced pyramid pooling transformer layer, which helps to strengthen the feature perception of the model. Experimental results demonstrate that AHTCN outperforms the state-of-the-art algorithms in terms of both subjective visual effects and objective quality assessment, while requiring fewer parameters.

show abstract

Section: Resultsmentioning

confidence: 94%

Section: Resultsmentioning

confidence: 99%

Section: Quantitative Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Attention-guided hybrid transformer-convolutional neural network for underwater image super-resolution

Zhan,

Li,

Zhang

2024

J. Electron. Imag.

View full text Add to dashboard Cite

show abstract

“…In more detail, HR images are downsampled by ×2, ×4, and ×8 scale factors, leveraging bicubic interpolation with 20% Gaussian noise. We utilize existing underwater SR methods for comparison, including EDSRGAN [31], SRGAN [26], SRResNet [26], DPSR [64], ESRGAN [55], SRDRM [19], SRDRM-GAN [19], PAL [5], and AMPCNet [67].…”

Section: Results On Underwater Imagementioning

confidence: 99%

Image Super-Resolution via Lightweight Attention-Directed Feature Aggregation Network

Wang

Tang

et al. 2023

ACM Trans. Multimedia Comput. Commun. Appl.

View full text Add to dashboard Cite

The advent of convolutional neural networks (CNN) has brought substantial progress in image super-resolution (SR) reconstruction. However, most SR methods pursue deep architectures to boost performance, and the resulting large model sizes are impractical for real-world applications. Furthermore, they insufficiently explore the internal structural information of image features, disadvantaging the restoration of fine texture details. To solve these challenges, we propose a lightweight architecture based on CNN named attention-directed feature aggregation network (AFAN), consisting of chained stacking multi-aware attention modules (MAAM) and a simple channel attention module (SCAM), for image SR. Specifically, in each MAAM, we construct a space-aware attention block (SAAB) and a dimension-aware attention block (DAAB) that individually yield unique three-dimensional modulation coefficients to adaptively recalibrate structural information from asymmetric convolution residual block (ACRB). The synergistic strategy captures multiple content features that are both space-aware and dimension-aware to preserve more fine-grained details. In addition, to further enhance the accuracy and robustness of the network, SCAM is embedded in the last MAAM to highlight channels with high activated values at low computational load. Comprehensive experiments verify that our proposed network attains high qualitative accuracy while employing fewer parameters and moderate computational requirements, exceeding most state-of-the-art lightweight approaches.

show abstract

Single-image super-resolution via a lightweight convolutional neural network with improved shuffle learning

Lu,

Xie,

et al. 2023

SIViP

View full text Add to dashboard Cite

Attention-guided multi-path cross-CNN for underwater image super-resolution

Cited by 16 publications

References 36 publications

Attention-guided hybrid transformer-convolutional neural network for underwater image super-resolution

Attention-guided hybrid transformer-convolutional neural network for underwater image super-resolution

Image Super-Resolution via Lightweight Attention-Directed Feature Aggregation Network

Single-image super-resolution via a lightweight convolutional neural network with improved shuffle learning

Contact Info

Product

Resources

About