A novel edge-feature attention fusion framework for underwater image enhancement DOI Creative Commons
Shuai Shen, Haoyi Wang, Weitao Chen

et al.

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 12

Published: April 4, 2025

Underwater images captured by Remotely Operated Vehicles are critical for marine research, ocean engineering, and national defense, but challenges such as blurriness color distortion necessitate advanced enhancement techniques. To address these issues, this paper presents the CUG-UIEF algorithm, an underwater image framework leveraging edge feature attention fusion. The method comprises three modules: 1) Attention-Guided Edge Feature Fusion Module that extracts information via operators enhances object detail through multi-scale integration with channel-cross to resolve blurring; 2) a Spatial Information Enhancement employs spatial-cross capture spatial interrelationships improve semantic representation, mitigating low signal-to-noise ratio; 3) Multi-Dimensional Perception Optimization integrating perceptual, structural, anomaly optimizations blurring contrast. Experimental results demonstrate achieves average peak ratio of 24.49 dB, 8.41% improvement over six mainstream algorithms, structural similarity index 0.92, 1.09% increase. These findings highlight model’s effectiveness in balancing preservation, semantics, perceptual quality, offering promising applications science related fields.

Language: Английский

Depthanything and SAM for UIE: exploring large model information contributes to underwater image restoration DOI
Jinxin Shao, Haosu Zhang, Jianming Miao

et al.

Machine Vision and Applications, Journal Year: 2025, Volume and Issue: 36(2)

Published: Feb. 11, 2025

Language: Английский

Citations

0
A novel edge-feature attention fusion framework for underwater image enhancement DOI Creative Commons
Shuai Shen, Haoyi Wang, Weitao Chen

et al.

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 12

Published: April 4, 2025

Underwater images captured by Remotely Operated Vehicles are critical for marine research, ocean engineering, and national defense, but challenges such as blurriness color distortion necessitate advanced enhancement techniques. To address these issues, this paper presents the CUG-UIEF algorithm, an underwater image framework leveraging edge feature attention fusion. The method comprises three modules: 1) Attention-Guided Edge Feature Fusion Module that extracts information via operators enhances object detail through multi-scale integration with channel-cross to resolve blurring; 2) a Spatial Information Enhancement employs spatial-cross capture spatial interrelationships improve semantic representation, mitigating low signal-to-noise ratio; 3) Multi-Dimensional Perception Optimization integrating perceptual, structural, anomaly optimizations blurring contrast. Experimental results demonstrate achieves average peak ratio of 24.49 dB, 8.41% improvement over six mainstream algorithms, structural similarity index 0.92, 1.09% increase. These findings highlight model’s effectiveness in balancing preservation, semantics, perceptual quality, offering promising applications science related fields.

Language: Английский

Citations

0