loading page

An Underwater Image Enhancement Model Combining Physical Priors and Residual Network
  • +2
  • Xinnan Fan,
  • Xuan Zhou,
  • Hongzhu Chen,
  • Yuanxue Xin,
  • Pengfei Shi
Xinnan Fan
Hohai University
Author Profile
Xuan Zhou
Hohai University - Changzhou Campus
Author Profile
Hongzhu Chen
Hohai University
Author Profile
Yuanxue Xin
College of Information Science and Engineering, Hohai University
Author Profile
Pengfei Shi
Hohai University College of Internet of Things Engineering

Corresponding Author:[email protected]

Author Profile

Abstract

Considering light absorbing and scattering problems in connection with wavelength can decrease the visibility, contrast and color distortion of images, we propose a new type of convolutional neural network with two training phases. Firstly, the coordinate attention module is integrated into the residual block of the residual group in the backbone network, which is used to strengthen the feature extraction capability of the network. Secondly, since the unrealistic image colors may degrade the image details, an unsupervised method that combines the physical prior knowledge and the real underwater images is proposed to finetune the backbone network. Furthermore, a model protection mechanism is designed to guarantee the successful execution of the training. The experimental results indicate the proposed model can effectively optimize the contrast, color and image quality of the underwater image. Compared with relevant algorithms, our UCIQE and NIQE are respectively 0.525 and 4.149, which further verifies the superiority of the proposed model.
12 Sep 2023Submitted to Electronics Letters
12 Sep 2023Submission Checks Completed
12 Sep 2023Assigned to Editor
14 Sep 2023Reviewer(s) Assigned
23 Sep 2023Review(s) Completed, Editorial Evaluation Pending
25 Sep 2023Editorial Decision: Revise Major
10 Oct 20231st Revision Received
17 Oct 2023Review(s) Completed, Editorial Evaluation Pending
18 Oct 2023Editorial Decision: Accept