Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Aberrations in Peripheral Blood Neuron-Specific Enolase Levels in Depression: A Systematic Review and Meta-Analysis.

Depression and anxiety·2026
Same author

IMATAC imputes single-cell ATAC-seq data by deep hierarchical network with denoising autoencoder.

Briefings in bioinformatics·2025
Same author

Letter to the editor regarding: "Slow gait speed is associated with frailty, activities of daily living and nutritional status in in-patient pulmonology patients."

Aging clinical and experimental research·2025
Same author

Letter to the Editor- "A machine-learning-derived online prediction model for depression risk in COPD patients: A retrospective cohort study from CHARLS".

Journal of affective disorders·2025
Same author

Association of dietary caffeine intake and metabolites in the urine of individuals with latent tuberculosis infection: a cross-sectional study.

BMC pulmonary medicine·2025
Same author

SurgRIPE challenge: Benchmark of surgical robot instrument pose estimation.

Medical image analysis·2025

Related Experiment Video

Updated: Jun 13, 2025

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging
09:19

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging

Published on: April 18, 2025

374

Lightweight deep learning model for underwater waste segmentation based on sonar images.

Yangke Li1, Xinman Zhang1

  • 1School of Automation Science and Engineering, Faculty of Electronic and Information Engineering, MOE Key Lab for Intelligent Networks and Network Security, Xi'an Jiaotong University, Xi'an 710049, Shaanxi, China.

Waste Management (New York, N.Y.)
|September 15, 2024
PubMed
Summary

This study introduces a lightweight network for segmenting underwater waste from sonar images, improving autonomous robot efficiency. The method enhances marine debris identification and recycling for sustainable development.

Keywords:
Cross-level interactionLightweight modelMulti-scale perceptionSemantic segmentationSonar imagesUnderwater waste

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.6K

Related Experiment Videos

Last Updated: Jun 13, 2025

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging
09:19

Measuring the Structure, Composition, and Change of Underwater Environments with Large-area Imaging

Published on: April 18, 2025

374
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.6K

Area of Science:

  • Marine Biology
  • Robotics
  • Computer Vision

Background:

  • Marine waste accumulation poses significant ecological threats and pollution challenges.
  • Traditional manual waste collection methods are inefficient and hazardous.
  • Automated underwater waste recycling is crucial for marine environmental protection.

Purpose of the Study:

  • To develop a lightweight, efficient network for underwater waste segmentation using sonar images.
  • To provide pixel-level location and category information for autonomous underwater robots.
  • To enhance the precision and efficiency of autonomous underwater waste recycling systems.

Main Methods:

  • Proposed a lightweight multi-scale cross-level network for sonar image segmentation.
  • Introduced hybrid perception and multi-scale attention modules for feature extraction.
  • Utilized sampling attention and cross-level interaction modules for feature fusion and down-sampling.

Main Results:

  • Achieved 74.66% mean Intersection over Union (mIoU) with only 0.68 million parameters.
  • Outperformed existing semantic segmentation models, including PIDNet Small and SeaFormer T.
  • Demonstrated significant improvements in mIoU (1.15% vs. PIDNet Small, 2.07% vs. SeaFormer T) with reduced parameters (91% fewer than PIDNet Small, 59% fewer than SeaFormer T).

Conclusions:

  • The proposed network offers a balanced performance between model size and segmentation accuracy for underwater waste.
  • This technology provides valuable insights for intelligent underwater waste recycling and promotes sustainable marine development.
  • The method enhances the capabilities of autonomous underwater robots in waste management.