Adaptive Deep Learning Network With Multi-Scale and Multi-Dimensional Features for Underwater Image Enhancement

Qiao, Nianzu; Dong, Lu; Sun, Changyin

doi:10.1109/tbc.2022.3227424

Cited by 8 publications

(2 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning (Hou et al 2023b) are widely used in image process tasks (Gui et al 2023). DL-UIE methods (Anil, Sreelatha et al 2023;Wang et al 2021;Sharma, Bisht, and Sur 2023;Qiao, Dong, and Sun 2022;Jiang et al 2022a;Liu et al 2022;Kang et al 2022) typically rely on underwater datasets with a sufficient number of images. An attention-guided dynamic multibranch neural network is designed by ADMNNet (Yan et al 2022), which is shown to be effective in extracting the multiscale features.…”

Section: Dl-uiementioning

confidence: 99%

Underwater Organism Color Fine-Tuning via Decomposition and Guidance

Cong,

Gui,

Hou

2024

AAAI

View full text Add to dashboard Cite

Due to the wavelength dependent light attenuation and scattering, the color of the underwater organism usually appears distorted. The existing underwater image enhancement methods mainly focus on designing networks capable of generating enhanced underwater organisms with fixed color. Due to the complexity of the underwater environment, ground truth labels are difficult to obtain, which results in the non-existence of perfect enhancement effects. Different from the existing methods, this paper proposes an algorithm with color enhancement and color fine-tuning (CECF) capabilities. The color enhancement behavior of CECF is the same as that of existing methods, aiming to restore the color of the distorted underwater organism. Beyond this general purpose, the color fine-tuning behavior of CECF can adjust the color of organisms in a controlled manner, which can generate enhanced organisms with diverse colors. To achieve this purpose, four processes are used in CECF. A supervised enhancement process learns the mapping from a distorted image to an enhanced image by the decomposition of color code. A self reconstruction process and a cross-reconstruction process are used for content-invariant learning. A color fine-tuning process is designed based on the guidance for obtaining various enhanced results with different colors. Experimental results have proven the enhancement ability and color fine-tuning ability of the proposed CECF. The source code is provided in https://github.com/Xiaofeng-life/CECF.

show abstract

Section: Dl-uiementioning

confidence: 99%

Underwater Organism Color Fine-Tuning via Decomposition and Guidance

Cong,

Gui,

Hou

2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Besides this, some researchers used the conventional methods: Gray World [41], Shades of Grey [42], Gray-Edge [43], max-RGB [44], [45] and Color Equalization [46] for removal of color casts in addition to restoration. The advancement of Deep Learning also provides a considerable contribution to underwater image enhancement, which include Deep residual net [47], Feedback Adversarial Transfer Learning [48], structure decomposition network [49], multiscale dense GAN [50], channel wise feature map based residual block [51], JLCL-Net [52], Feature Fusion Network [53], UW-GAN [54], TOPAL [55], DPIENet [56], SCEIR [57], SGUIE-Net [58], UAGAN [59], Fusion of features of RGB and HSV spaces by MSDC-Net [60], designing of neural network using physical model of [6] and [7] by HybrUR [61], adaptive DLN with dewater pooling and new attention mechanism [62], U-shape transformer was integrated with the combination of channel-wise and global-wise modelling transformer [63]. Underwater image curve model was proposed based on model of haze image for the enhancement of degraded images without any reference images (Zero-UIE) [64] where lightweight deep network was used for the estimation of curve parameters.…”

Section: Introductionmentioning

confidence: 99%

Adjustment of Pixel Intensity for Removal of Color Casts in Underwater Images

Kanagavel,

Thanikaiselvan

2024

IEEE Access

View full text Add to dashboard Cite

This article introduces a new color correction method based on the adjustment of pixel intensity to remove color casts in underwater images. As the underwater images are captured in non-homogeneous with nonuniform illumination, the Gray-World Assumption (GWA) causes red artefacts. The method mainly aims to overcome the drawback of GWA for white balancing the underwater images. Before employing GWA, the underwater image is preprocessed by adding a color inferiority-reducing factor to the inferior channels to increase the pixel intensity level. The GWA is applied to preserve the information for substantially removing color casts. The image is adjusted by saturating the 1% top and bottom intensity levels pixels to remove the water haze. The color dominancy balanced image is finetuned by multi-scale image fusion technique with three derived inputs. The proposed method outperforms qualitatively and quantitatively based on the comparative analysis with four existing methods.

show abstract

Diving into Clarity: Restoring Underwater Images using Deep Learning

Martinho,

Calvalcanti,

Pio

et al. 2024

J Intell Robot Syst

View full text Add to dashboard Cite

In this paper we propose a learning-based restoration approach to learn the optimal parameters for enhancing the quality of different types of underwater images and apply a set of intensity transformation techniques to process raw underwater images. The methodology comprises two steps. Firstly, a Convolutional Neural Network (CNN) Regression model is employed to learn enhancing parameters for each underwater image type. Trained on a diverse dataset, the CNN captures complex relationships, enabling generalization to various underwater conditions. Secondly, we apply intensity transformation techniques to raw underwater images. These transformations collectively compensate for visual information loss due to underwater degradation, enhancing overall image quality. In order to evaluate the performance of our proposed approach, we conducted qualitative and quantitative experiments using well-known underwater image datasets (U45 and UIEB), and using the proposed challenging dataset composed by 276 underwater images from the Amazon region (AUID). The results demonstrate that our approach achieves an impressive accuracy rate in different underwater image datasets. For U45 and UIEB datasets, regarding PSNR and SSIM quality metrics, we achieved 26.967, 0.847, 27.299 and 0.793, respectively. Meanwhile, the best comparison techniques achieved 26.879, 0.831, 27.157 and 0.788, respectively.

show abstract

Adaptive Deep Learning Network With Multi-Scale and Multi-Dimensional Features for Underwater Image Enhancement

Cited by 8 publications

References 38 publications

Underwater Organism Color Fine-Tuning via Decomposition and Guidance

Underwater Organism Color Fine-Tuning via Decomposition and Guidance

Adjustment of Pixel Intensity for Removal of Color Casts in Underwater Images

Diving into Clarity: Restoring Underwater Images using Deep Learning

Contact Info

Product

Resources

About