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

Extraction: Advanced Methods00:56

Extraction: Advanced Methods

1.3K
Metal ions can be separated from one another by complexation with organic ligands–the chelating agent– to form uncharged chelates. Here, the chelating agent must contain hydrophobic groups and behave as a weak acid, losing a proton to bind with the metal. Since most organic ligands used in this process are insoluble or undergo oxidation in the aqueous phase, the chelating agent is initially added to the organic phase and extracted into the aqueous phase. The metal-ligand complex is...
1.3K
Parallel Processing01:20

Parallel Processing

950
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
950

You might also read

Related Articles

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

Sort by
Same author

Automated Landslide Identification from Time-Series InSAR Using Improved Hot Spot Analysis.

Sensors (Basel, Switzerland)·2026
Same author

Polygonatum sibiricum polysaccharide delays age-related decline and preserves muscle integrity in Caenorhabditis elegans by modulating the daf-2/IIS pathway.

International journal of biological macromolecules·2026
Same author

Establishment and Validation of a Competency-Based Evaluation Indicator System for On-Site Disinfection Service Providers - Shanghai Municipality, China, 2023.

China CDC weekly·2026
Same author

Iron overload disrupts lipid desaturation through SKN-1 signaling in <i>C. elegans</i>: implications for ferroptosis-based immunomodulatory biomaterials.

Materials today. Bio·2026
Same author

Upcycling Mixed Alloy Scraps via Liquid Metal Dealloying for Cross-Alloy Circular Economy.

Advanced materials (Deerfield Beach, Fla.)·2026
Same author

Copper overload induces lifespan shortening through activating mitochondrial respiration in Caenorhabditis elegans.

Journal of hazardous materials·2026

Related Experiment Video

Updated: May 3, 2026

Time Multiplexing Super Resolving Technique for Imaging from a Moving Platform
06:25

Time Multiplexing Super Resolving Technique for Imaging from a Moving Platform

Published on: February 12, 2014

8.5K

A Building Extraction Method for High-Resolution Remote Sensing Images with Multiple Attentions and Parallel Encoders

Zhaojun Pang1, Rongming Hu1, Wu Zhu2

  • 1School of Geomatics, Xi'an University of Science and Technology, Xi'an 710054, China.

Sensors (Basel, Switzerland)
|February 10, 2024
PubMed
Summary

This study introduces MARS-Net, a novel deep learning model for accurate pixel-level building extraction from remote sensing images. The model achieves superior performance by integrating parallel encoders, attention mechanisms, and spectral enhancement for improved building segmentation.

Keywords:
building extractiondeep convolutional neural network (DCNN)high-resolution remote sensing imageryspectral enhancementtransformer

More Related Videos

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

7.6K
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

475

Related Experiment Videos

Last Updated: May 3, 2026

Time Multiplexing Super Resolving Technique for Imaging from a Moving Platform
06:25

Time Multiplexing Super Resolving Technique for Imaging from a Moving Platform

Published on: February 12, 2014

8.5K
A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

7.6K
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

475

Area of Science:

  • Remote Sensing
  • Computer Vision
  • Geographical Information Systems

Background:

  • Accurate pixel-level building extraction from high-resolution remote sensing images is crucial for geospatial applications.
  • Variations in building shapes, distributions, and complex spectral features across different regions pose challenges for consistent segmentation.
  • Existing methods struggle to maintain stable segmentation effects and capture multi-scale building details effectively.

Purpose of the Study:

  • To develop an advanced deep learning model for robust and accurate pixel-level building extraction from remote sensing data.
  • To address the challenges posed by diverse building appearances and complex spectral characteristics in remote sensing imagery.
  • To improve the segmentation performance and detail preservation of buildings in varying geographical contexts.

Main Methods:

  • A parallel encoded building extraction network (MARS-Net) was proposed, utilizing both Deep Convolutional Neural Networks (DCNN) and Transformers for local and global feature extraction.
  • Incorporated Coordinate Attention (CA) and Convolutional Block Attention Module (CBAM) to enhance spatial and semantic information transfer between the encoder and decoder.
  • Integrated Dense Atrous Spatial Pyramid Pooling (DenseASPP) for multi-scale contextual information capture and a Spectral Information Enhancement Module (SIEM) for multi-band data fusion and enhancement.

Main Results:

  • MARS-Net demonstrated superior building extraction performance compared to existing methods.
  • The inclusion of the Spectral Information Enhancement Module (SIEM) further improved segmentation accuracy and detail.
  • Achieved an IoU of 87.53% on the Xi'an dataset and 89.62% on the WHU dataset, with F1 scores of 93.34% and 94.52%, respectively.

Conclusions:

  • MARS-Net offers a significant advancement in pixel-level building extraction from high-resolution remote sensing images.
  • The proposed attention mechanisms and spectral enhancement module effectively address challenges related to building variability and spectral complexity.
  • The model's high IoU and F1 scores validate its effectiveness for practical geospatial information applications.