MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequency
IntroductionThe advancement of Underwater Human-Robot Interaction technology has significantly driven marine exploration, conservation, and resource utilization. However, challenges persist due to the limitations of underwater robots equipped with basic cameras, which struggle to handle complex unde...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Marine Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmars.2025.1541265/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823859415477911552 |
|---|---|
| author | Dechuan Kong Dechuan Kong Yandi Zhang Xiaohu Zhao Yanqiang Wang Lei Cai |
| author_facet | Dechuan Kong Dechuan Kong Yandi Zhang Xiaohu Zhao Yanqiang Wang Lei Cai |
| author_sort | Dechuan Kong |
| collection | DOAJ |
| description | IntroductionThe advancement of Underwater Human-Robot Interaction technology has significantly driven marine exploration, conservation, and resource utilization. However, challenges persist due to the limitations of underwater robots equipped with basic cameras, which struggle to handle complex underwater environments. This leads to blurry images, severely hindering the performance of automated systems.MethodsWe propose MUFFNet, an underwater image enhancement network leveraging multi-scale frequency analysis to address the challenge. The network introduces a frequency-domain-based convolutional attention mechanism to extract spatial information effectively. A Multi-Scale Enhancement Prior algorithm enhances high-frequency and low-frequency features while the Information Flow Interaction module mitigates information stratification and blockage. A Multi-Scale Joint Loss framework facilitates dynamic network optimization.ResultsExperimental results demonstrate that MUFFNet outperforms existing state-of-the-art models while consuming fewer computational resources and aligning enhanced images more closely with human visual perception.DiscussionThe enhanced images generated by MUFFNet exhibit better alignment with human visual perception, making it a promising solution for improving underwater robotic vision systems. |
| format | Article |
| id | doaj-art-a0184beceb3c440bb2e20c35c3a15e15 |
| institution | Kabale University |
| issn | 2296-7745 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Marine Science |
| spelling | doaj-art-a0184beceb3c440bb2e20c35c3a15e152025-02-11T05:10:26ZengFrontiers Media S.A.Frontiers in Marine Science2296-77452025-02-011210.3389/fmars.2025.15412651541265MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequencyDechuan Kong0Dechuan Kong1Yandi Zhang2Xiaohu Zhao3Yanqiang Wang4Lei Cai5School of Artificial Intelligence, Henan Institute of Science and Technology, Xinxiang, ChinaNational and Local Joint Engineering Laboratory of Internet Application Technology on Mine, China University of Mining and Technology, Xuzhou, ChinaSchool of Information Science and Engineering, Shenyang University of Technology, Shenyang, ChinaNational and Local Joint Engineering Laboratory of Internet Application Technology on Mine, China University of Mining and Technology, Xuzhou, ChinaSchool of Artificial Intelligence, Henan Institute of Science and Technology, Xinxiang, ChinaSchool of Artificial Intelligence, Henan Institute of Science and Technology, Xinxiang, ChinaIntroductionThe advancement of Underwater Human-Robot Interaction technology has significantly driven marine exploration, conservation, and resource utilization. However, challenges persist due to the limitations of underwater robots equipped with basic cameras, which struggle to handle complex underwater environments. This leads to blurry images, severely hindering the performance of automated systems.MethodsWe propose MUFFNet, an underwater image enhancement network leveraging multi-scale frequency analysis to address the challenge. The network introduces a frequency-domain-based convolutional attention mechanism to extract spatial information effectively. A Multi-Scale Enhancement Prior algorithm enhances high-frequency and low-frequency features while the Information Flow Interaction module mitigates information stratification and blockage. A Multi-Scale Joint Loss framework facilitates dynamic network optimization.ResultsExperimental results demonstrate that MUFFNet outperforms existing state-of-the-art models while consuming fewer computational resources and aligning enhanced images more closely with human visual perception.DiscussionThe enhanced images generated by MUFFNet exhibit better alignment with human visual perception, making it a promising solution for improving underwater robotic vision systems.https://www.frontiersin.org/articles/10.3389/fmars.2025.1541265/fullunderwater image enhancementunderwater human-robot interactionmulti-scale knowledgemulti-frequency extractionconvolutional attentiondeep learning |
| spellingShingle | Dechuan Kong Dechuan Kong Yandi Zhang Xiaohu Zhao Yanqiang Wang Lei Cai MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequency Frontiers in Marine Science underwater image enhancement underwater human-robot interaction multi-scale knowledge multi-frequency extraction convolutional attention deep learning |
| title | MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequency |
| title_full | MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequency |
| title_fullStr | MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequency |
| title_full_unstemmed | MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequency |
| title_short | MUFFNet: lightweight dynamic underwater image enhancement network based on multi-scale frequency |
| title_sort | muffnet lightweight dynamic underwater image enhancement network based on multi scale frequency |
| topic | underwater image enhancement underwater human-robot interaction multi-scale knowledge multi-frequency extraction convolutional attention deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fmars.2025.1541265/full |
| work_keys_str_mv | AT dechuankong muffnetlightweightdynamicunderwaterimageenhancementnetworkbasedonmultiscalefrequency AT dechuankong muffnetlightweightdynamicunderwaterimageenhancementnetworkbasedonmultiscalefrequency AT yandizhang muffnetlightweightdynamicunderwaterimageenhancementnetworkbasedonmultiscalefrequency AT xiaohuzhao muffnetlightweightdynamicunderwaterimageenhancementnetworkbasedonmultiscalefrequency AT yanqiangwang muffnetlightweightdynamicunderwaterimageenhancementnetworkbasedonmultiscalefrequency AT leicai muffnetlightweightdynamicunderwaterimageenhancementnetworkbasedonmultiscalefrequency |