English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Underwater image enhancement (UIE) is a challenging task due to the complexdegradation caused by underwater environments. To solve this issue, previousmethods often idealize the degradation process, and neglect the impact ofmedium noise and object motion on the distribution of image features, limitingthe generalization and adaptability of the model. Previous methods use thereference gradient that is constructed from original images and syntheticground-truth images. This may cause the network performance to be influenced bysome low-quality training data. Our approach utilizes predicted images todynamically update pseudo-labels, adding a dynamic gradient to optimize thenetwork's gradient space. This process improves image quality and avoids localoptima. Moreover, we propose a Feature Restoration and Reconstruction module(FRR) based on a Channel Combination Inference (CCI) strategy and a FrequencyDomain Smoothing module (FRS). These modules decouple other degradationfeatures while reducing the impact of various types of noise on networkperformance. Experiments on multiple public datasets demonstrate thesuperiority of our method over existing state-of-the-art approaches, especiallyin achieving performance milestones: PSNR of 25.6dB and SSIM of 0.93 on theUIEB dataset. Its efficiency in terms of parameter size and inference timefurther attests to its broad practicality. The code will be made publiclyavailable.

DGNet: Dynamic Gradient-guided Network with Noise Suppression for Underwater Image Enhancement