U-TransCNN: A U-shape transformer-CNN fusion model for underwater image enhancement

Yao Haiyang; Guo Ruige; Zhao Zhongda; Zang Yuzhang; Zhao Xiaobo; Lei Tao; Wang Haiyan

doi:10.1016/j.displa.2025.103047

U-TransCNN: A U-shape transformer-CNN fusion model for underwater image enhancement

Yao Haiyang^*, Guo Ruige, Zhao Zhongda, Zang Yuzhang, Zhao Xiaobo, Lei Tao, Wang Haiyan

^*Corresponding author for this work

Department of Electrical and Computer Engineering - Communication, Control and Automation

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Abstract

Underwater imaging faces significant challenges due to nonuniform optical absorption and scattering, resulting in visual quality issues like color distortion, contrast reduction, and image blurring. These factors hinder the accurate capture and clear depiction of underwater imagery. To address these complexities, we propose U-TransCNN, a U-shape Transformer- Convolutional Neural Networks (CNN) model, designed to enhance underwater images by integrating the strengths of CNNs and Transformers. The core of U-TransCNN is the Global-Detail Feature Synchronization Fusion Module. This innovative component enhances global color and contrast while meticulously preserving the intricate texture details, ensuring that both macroscopic and microscopic aspects of the image are enhanced in unison. Then we design the Multiscale Detail Fusion Block to aggregate a richer spectrum of feature information using a variety of convolution kernels. Furthermore, our optimization strategy is augmented with a joint loss function, adynamic approach allowing the model to assign varying weights to the loss associated with different pixel points, depending on their loss magnitude. Six experiments (including reference and non-reference) on three public underwater datasets confirm that U-TransCNN comprehensively surpasses other contemporary state-of-the-art deep learning algorithms, demonstrating marked improvement in visualization quality and quantization parameters of underwater images. Our code is available at https://github.com/GuoRuige/UTransCNN.

Original language	English
Article number	103047
Journal	Displays
Volume	88
ISSN	0141-9382
DOIs	https://doi.org/10.1016/j.displa.2025.103047
Publication status	Published - Jul 2025

Keywords

CNN
Feature fusion
Transformer
Underwater image enhancement

Access to Document

10.1016/j.displa.2025.103047

Cite this

@article{e51794304d3748a98669555273a5baac,

title = "U-TransCNN: A U-shape transformer-CNN fusion model for underwater image enhancement",

abstract = "Underwater imaging faces significant challenges due to nonuniform optical absorption and scattering, resulting in visual quality issues like color distortion, contrast reduction, and image blurring. These factors hinder the accurate capture and clear depiction of underwater imagery. To address these complexities, we propose U-TransCNN, a U-shape Transformer- Convolutional Neural Networks (CNN) model, designed to enhance underwater images by integrating the strengths of CNNs and Transformers. The core of U-TransCNN is the Global-Detail Feature Synchronization Fusion Module. This innovative component enhances global color and contrast while meticulously preserving the intricate texture details, ensuring that both macroscopic and microscopic aspects of the image are enhanced in unison. Then we design the Multiscale Detail Fusion Block to aggregate a richer spectrum of feature information using a variety of convolution kernels. Furthermore, our optimization strategy is augmented with a joint loss function, adynamic approach allowing the model to assign varying weights to the loss associated with different pixel points, depending on their loss magnitude. Six experiments (including reference and non-reference) on three public underwater datasets confirm that U-TransCNN comprehensively surpasses other contemporary state-of-the-art deep learning algorithms, demonstrating marked improvement in visualization quality and quantization parameters of underwater images. Our code is available at https://github.com/GuoRuige/UTransCNN.",

keywords = "CNN, Feature fusion, Transformer, Underwater image enhancement",

author = "Yao Haiyang and Guo Ruige and Zhao Zhongda and Zang Yuzhang and Zhao Xiaobo and Lei Tao and Wang Haiyan",

note = "Publisher Copyright: {\textcopyright} 2025 Elsevier B.V.",

year = "2025",

month = jul,

doi = "10.1016/j.displa.2025.103047",

language = "English",

volume = "88",

journal = "Displays",

issn = "0141-9382",

publisher = "Elsevier BV",

}

TY - JOUR

T1 - U-TransCNN

T2 - A U-shape transformer-CNN fusion model for underwater image enhancement

AU - Haiyang, Yao

AU - Ruige, Guo

AU - Zhongda, Zhao

AU - Yuzhang, Zang

AU - Xiaobo, Zhao

AU - Tao, Lei

AU - Haiyan, Wang

PY - 2025/7

Y1 - 2025/7

N2 - Underwater imaging faces significant challenges due to nonuniform optical absorption and scattering, resulting in visual quality issues like color distortion, contrast reduction, and image blurring. These factors hinder the accurate capture and clear depiction of underwater imagery. To address these complexities, we propose U-TransCNN, a U-shape Transformer- Convolutional Neural Networks (CNN) model, designed to enhance underwater images by integrating the strengths of CNNs and Transformers. The core of U-TransCNN is the Global-Detail Feature Synchronization Fusion Module. This innovative component enhances global color and contrast while meticulously preserving the intricate texture details, ensuring that both macroscopic and microscopic aspects of the image are enhanced in unison. Then we design the Multiscale Detail Fusion Block to aggregate a richer spectrum of feature information using a variety of convolution kernels. Furthermore, our optimization strategy is augmented with a joint loss function, adynamic approach allowing the model to assign varying weights to the loss associated with different pixel points, depending on their loss magnitude. Six experiments (including reference and non-reference) on three public underwater datasets confirm that U-TransCNN comprehensively surpasses other contemporary state-of-the-art deep learning algorithms, demonstrating marked improvement in visualization quality and quantization parameters of underwater images. Our code is available at https://github.com/GuoRuige/UTransCNN.

AB - Underwater imaging faces significant challenges due to nonuniform optical absorption and scattering, resulting in visual quality issues like color distortion, contrast reduction, and image blurring. These factors hinder the accurate capture and clear depiction of underwater imagery. To address these complexities, we propose U-TransCNN, a U-shape Transformer- Convolutional Neural Networks (CNN) model, designed to enhance underwater images by integrating the strengths of CNNs and Transformers. The core of U-TransCNN is the Global-Detail Feature Synchronization Fusion Module. This innovative component enhances global color and contrast while meticulously preserving the intricate texture details, ensuring that both macroscopic and microscopic aspects of the image are enhanced in unison. Then we design the Multiscale Detail Fusion Block to aggregate a richer spectrum of feature information using a variety of convolution kernels. Furthermore, our optimization strategy is augmented with a joint loss function, adynamic approach allowing the model to assign varying weights to the loss associated with different pixel points, depending on their loss magnitude. Six experiments (including reference and non-reference) on three public underwater datasets confirm that U-TransCNN comprehensively surpasses other contemporary state-of-the-art deep learning algorithms, demonstrating marked improvement in visualization quality and quantization parameters of underwater images. Our code is available at https://github.com/GuoRuige/UTransCNN.

KW - CNN

KW - Feature fusion

KW - Transformer

KW - Underwater image enhancement

UR - http://www.scopus.com/inward/record.url?scp=105001480655&partnerID=8YFLogxK

U2 - 10.1016/j.displa.2025.103047

DO - 10.1016/j.displa.2025.103047

M3 - Journal article

AN - SCOPUS:105001480655

SN - 0141-9382

VL - 88

JO - Displays

JF - Displays

M1 - 103047

ER -

U-TransCNN: A U-shape transformer-CNN fusion model for underwater image enhancement

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this