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

A Dicerium(IV) Nitride Complex with a Linear CeNCe Core Stabilized by Bulky Cyclohexyltriamide Ligands.

Journal of the American Chemical Society·2026
Same author

Multi-Tissue Transcriptomic Analysis Reveals Tissue-Specific Thermal Responses and Sex-Biased Functional Differentiation in Reproductive Tissues of the Brown Frog (<i>Rana dybowskii</i>).

Current issues in molecular biology·2026
Same author

Postoperative changes in cervical hemodynamics and cognitive function following cervical lymphatic-venous surgery in Alzheimer's disease.

Journal of Alzheimer's disease : JAD·2026
Same author

Ozonated water attenuates nucleus pulposus cell apoptosis and matrix degradation by suppressing NF-κB signaling in a rat model of <i>Staphylococcus aureus</i>-induced discitis.

Connective tissue research·2026
Same author

Upfront triple combination therapy with selexipag: insights from a real world cohort in Chinese patients with pulmonary arterial hypertension.

Frontiers in cardiovascular medicine·2026
Same author

UE, CT, and ACR TI-RADS Grading in the Combined Diagnosis of Benign and Malignant Thyroid Nodules.

APMIS : acta pathologica, microbiologica, et immunologica Scandinavica·2026

Related Experiment Video

Updated: Jun 9, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

370

RTINet: A Lightweight and High-Performance Railway Turnout Identification Network Based on Semantic Segmentation.

Dehua Wei1,2, Wenjun Zhang3, Haijun Li1,2

  • 1School of Traffic and Transportation, Lanzhou Jiaotong University, Lanzhou 730070, China.

Entropy (Basel, Switzerland)
|October 25, 2024
PubMed
Summary

A new railway turnout identification method using semantic segmentation improves train safety and driver workload. The RTINet model achieves high accuracy and speed for real-time railway applications.

Keywords:
Deeplabv3+attention mechanismlightweight networkrailway turnout identificationsemantic segmentation

More Related Videos

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

1.3K
Inherent Dynamics Visualizer, an Interactive Application for Evaluating and Visualizing Outputs from a Gene Regulatory Network Inference Pipeline
10:44

Inherent Dynamics Visualizer, an Interactive Application for Evaluating and Visualizing Outputs from a Gene Regulatory Network Inference Pipeline

Published on: December 7, 2021

2.1K

Related Experiment Videos

Last Updated: Jun 9, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

370
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

1.3K
Inherent Dynamics Visualizer, an Interactive Application for Evaluating and Visualizing Outputs from a Gene Regulatory Network Inference Pipeline
10:44

Inherent Dynamics Visualizer, an Interactive Application for Evaluating and Visualizing Outputs from a Gene Regulatory Network Inference Pipeline

Published on: December 7, 2021

2.1K

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Railway Engineering

Background:

  • Railway transportation safety is paramount, requiring efficient systems for train operations.
  • Current methods for identifying railway turnouts may not meet the demands of high-speed rail safety and automation.
  • Train driver workload can be reduced through intelligent automation of critical identification tasks.

Purpose of the Study:

  • To develop an intelligent semantic segmentation method for accurate railway turnout identification.
  • To enhance railway transportation safety and reduce the workload of train drivers.
  • To propose a novel railway turnout identification network (RTINet) suitable for real-time deployment.

Main Methods:

  • Construction and manual annotation of a railway turnout scene perception (RTSP) dataset, including side rails.
  • Development of the RTINet model based on Deeplabv3+, utilizing MobileNetV2 as a lightweight backbone.
  • Integration of depth-separable convolutions and the bottleneck attention module (BAM) for efficiency and enhanced perception.
  • Application of the Dice loss function to address foreground-background class imbalance during training.

Main Results:

  • The proposed RTINet achieved a mean Intersection over Union (mIoU) of 85.94% on the customized dataset.
  • The model demonstrated a fast inference speed of 78 frames per second (fps), suitable for high-speed trains.
  • Experimental results showed that RTINet outperformed existing baseline models in railway turnout identification.
  • An ablation study confirmed the effectiveness of individual optimized components within the RTINet architecture.

Conclusions:

  • The developed semantic segmentation method provides a feasible and effective solution for railway turnout identification.
  • The RTINet model offers a robust and efficient approach for enhancing railway safety and operational efficiency.
  • The study highlights the potential of lightweight deep learning models with attention mechanisms in safety-critical railway applications.