Search research articles

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

Dynamic bidirectional data recomposition for efficient road garbage segmentation in semi-supervised learning.

Neural networks : the official journal of the International Neural Network Society·2026

Same author

U<sup>2</sup>-Net and ResNet50-Based Automatic Pipeline for Bacterial Colony Counting.

Microorganisms·2024

Same author

Research on system of ultra-flat carrying robot based on improved PSO algorithm.

Frontiers in neurorobotics·2023

Same author

Spatiotemporal Feature Enhancement Aids the Driving Intention Inference of Intelligent Vehicles.

International journal of environmental research and public health·2022

Same author

Integrated strategy for widely targeted metabolome characterization of Peucedani Radix.

Journal of chromatography. A·2022

Same author

Simultaneous determination of eight tryptic peptides in musk using high-performance liquid chromatography coupled with tandem mass spectrometry.

Journal of chromatography. B, Analytical technologies in the biomedical and life sciences·2021

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Dec 6, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

UnetDVH-Linear: Linear Feature Segmentation by Dilated Convolution with Vertical and Horizontal Kernels.

Jiacai Liao¹, Libo Cao¹, Wei Li¹

¹State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410006, China.

Sensors (Basel, Switzerland)

|October 14, 2020

Summary

This study introduces dilated convolution with vertical and horizontal kernels (DVH) to improve semantic segmentation for linear features like lanes. The new method enhances neural networks for better linear feature extraction and segmentation accuracy.

Keywords:

dilated convolution linear features neutral networks semantic segmentation

More Related Videos

Quantifying Intermembrane Distances with Serial Image Dilations

Quantifying Intermembrane Distances with Serial Image Dilations

Published on: September 28, 2018

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Related Experiment Videos

Last Updated: Dec 6, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Quantifying Intermembrane Distances with Serial Image Dilations

Quantifying Intermembrane Distances with Serial Image Dilations

Published on: September 28, 2018

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Area of Science:

Computer Vision
Deep Learning
Image Segmentation

Background:

Extracting linear features like lanes and road markings is challenging for semantic segmentation networks due to their global contextual dependency.
Existing methods struggle to capture complete information for these elongated objects.

Purpose of the Study:

To enhance the linear feature extraction capabilities of semantic segmentation networks.
To investigate the impact of integrating dilated convolution with specifically designed vertical and horizontal kernels (DVH) into feature extraction.

Main Methods:

Proposed a novel approach by introducing dilated convolution with vertical and horizontal kernels (DVH) into semantic segmentation networks.
Evaluated the method on the SS dataset, TuSimple lane dataset, and Massachusetts Roads dataset.
Analyzed the performance of DVH kernels placed at different network locations.

Main Results:

Achieved a 2% accuracy improvement in slot marking segmentation on the SS dataset.
The UnetDVH-Linear (v1) model reached 97.53% accuracy on the TuSimple Benchmark Lane Detection Challenge.
Demonstrated strong generalization with 95.3% segmentation accuracy on the Massachusetts roads dataset without data augmentation.

Conclusions:

Dilated convolution with vertical and horizontal kernels (DVH) effectively enhances neural network performance in linear feature extraction.
The proposed method outperforms state-of-the-art approaches in lane and road segmentation tasks.
DVH integration offers a robust solution for improving semantic segmentation of linear objects.