App. Res. of Comp. Accepted Paper
Vol. 43, 2026 No. 12

Multi-label classification network for underwater video based on quality-guided Fusion mechanism

Li Yun1,4
Fang Menghao1,5
Pang Guangyao2
Wei Junfeng1,7
Qin Yuhua1,6
Jing Peiguang3
1. School of Physics and Electronic Information, Guangxi Minzu University, Nanning Guangxi 530000, China
2. Universities Key Laboratory of Intelligent Software, Wuzhou University, Wuzhou Guangxi 543002, China
3. School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
4. Guangxi Yuanjing Symbiotic Embodied Intelligence Science and Technology Innovation Cooperation Base, Nanning Guangxi 530000, China
5. Guangxi Engineering Research Center of Intelligent Vision and Collaborative Robotics, Nanning Guangxi 530000, China
6. Guangxi Key Laboratory of ZHIYU Humanoid Robots, Nanning Guangxi 530000, China
7. Engineering Research Center of Multi-modal Information Intelligent Sensing, Processing and Application, University of Nanning Guangxi 530000, China
DOI: 10.19734/j.issn.1001-3695.2026.02.0091

Abstract

Underwater videos suffer from degraded visual features, which increases the difficulty of multimodal semantic understanding. Existing video classification methods rely heavily on high-quality visual conditions. Multi-view fusion suffers from spatial misalignment and semantic shift, while multi-modal fusion lacks explicit modeling of feature quality differences. To address these problems, this paper proposed a multi-label classification network for underwater video based on a quality-guided fusion mechanism, named MQGNet. The network used an Adaptive Feature Alignment module (AFA) to achieve feature alignment and complementary fusion between original images and enhanced images. It employed a Quality-guided Multimodal Cross-Attention mechanism (QMC) to use textual semantics as high-quality semantic cues to guide low-quality visual features. It also introduced a Self-supervised Generative Adversarial Network module (SGAN) to restore visual features in complex underwater degradation scenarios. Experimental results showed that the proposed method achieved a mean Average Precision (mAP) of 0.8760 on the self-constructed Underwater Video-Text Dataset (UVT) . Its classification performance surpassed existing mainstream methods. The method also demonstrated good generalization ability on public datasets. These results verify the effectiveness and robustness of the proposed method in underwater video multi-label classification tasks.

Foundation Support

广西民族大学引进人才科研启动项目(2024KJQD218)
国家自然科学基金资助项目(62361002)
广西自然科学基金资助项目(2025GXNSFDA04240009)

Publish Information

DOI: 10.19734/j.issn.1001-3695.2026.02.0091
Publish at: Application Research of Computers Accepted Paper, Vol. 43, 2026 No. 12

Publish History

[2026-07-29] Accepted Paper

Cite This Article

李云, 方梦豪, 庞光垚, 等. 基于质量引导融合机制的水下视频多标签分类网络 [J]. 计算机应用研究, 2026, 43 (12). (2026-08-25). https://doi.org/10.19734/j.issn.1001-3695.2026.02.0091. (Li Yun, Fang Menghao, Pang Guangyao, et al. Multi-label classification network for underwater video based on quality-guided Fusion mechanism [J]. Application Research of Computers, 2026, 43 (12). (2026-08-25). https://doi.org/10.19734/j.issn.1001-3695.2026.02.0091. )

About the Journal

  • Application Research of Computers Monthly Journal
  • Journal ID ISSN 1001-3695
    CN  51-1196/TP

Application Research of Computers, founded in 1984, is an academic journal of computing technology sponsored by Sichuan Institute of Computer Sciences under the Science and Technology Department of Sichuan Province.

Aiming at the urgently needed cutting-edge technology in this discipline, Application Research of Computers reflects the mainstream technology, hot technology and the latest development trend of computer application research at home and abroad in a timely manner. The main contents of the journal include high-level academic papers in this discipline, the latest scientific research results and major application results. The contents of the columns involve new theories of computer discipline, basic computer theory, algorithm theory research, algorithm design and analysis, blockchain technology, system software and software engineering technology, pattern recognition and artificial intelligence, architecture, advanced computing, parallel processing, database technology, computer network and communication technology, information security technology, computer image graphics and its latest hot application technology.

Application Research of Computers has many high-level readers and authors, and its readers are mainly senior and middle-level researchers and engineers engaged in the field of computer science, as well as teachers and students majoring in computer science and related majors in colleges and universities. Over the years, the total citation frequency and Web download rate of Application Research of Computers have been ranked among the top of similar academic journals in this discipline, and the academic papers published are highly popular among the readers for their novelty, academics, foresight, orientation and practicality.

Indexed & Evaluation

  • The Second National Periodical Award 100 Key Journals
  • Double Effect Journal of China Journal Formation
  • the Core Journal of China (Peking University 2023 Edition)
  • the Core Journal for Science
  • Chinese Science Citation Database (CSCD) Source Journals
  • RCCSE Chinese Core Academic Journals
  • Journal of China Computer Federation
  • 2020-2022 The World Journal Clout Index (WJCI) Report of Scientific and Technological Periodicals
  • Full-text Source Journal of China Science and Technology Periodicals Database
  • Source Journal of China Academic Journals Comprehensive Evaluation Database
  • Source Journals of China Academic Journals (CD-ROM Version), China Journal Network
  • 2017-2019 China Outstanding Academic Journals with International Influence (Natural Science and Engineering Technology)
  • Source Journal of Top Academic Papers (F5000) Program of China's Excellent Science and Technology Journals
  • Source Journal of China Engineering Technology Electronic Information Network and Electronic Technology Literature Database
  • Source Journal of British Science Digest (INSPEC)
  • Japan Science and Technology Agency (JST) Source Journal
  • Russian Journal of Abstracts (AJ, VINITI) Source Journals
  • Full-text Journal of EBSCO, USA
  • Cambridge Scientific Abstracts (Natural Sciences) (CSA(NS)) core journals
  • Poland Copernicus Index (IC)
  • Ulrichsweb (USA)