The strong substructure and feature attention mechanism for image semantic segmentation

Semantic segmentation is a hot topic in computer vision and various deep learning networks are designed to achieve higher accuracy on that by fully exploring the capability of neural networks. This paper aims to address the issue and proposes the substructures with novelty for popular networks. Mean...

Full description

Saved in:
Bibliographic Details
Published inConcurrency and computation Vol. 34; no. 12
Main Authors Zhang, Yuhang, Ren, Hongshuai, Yang, Wensi, Wang, Yang, Ye, Kejiang, Xu, Cheng‐Zhong
Format Journal Article
LanguageEnglish
Published Hoboken, USA John Wiley & Sons, Inc 30.05.2022
Wiley Subscription Services, Inc
Subjects
a cross‐channel structure
Computer vision
Datasets
Generative adversarial networks
Image enhancement
Image segmentation
Neural networks
Object recognition
satellite image recognition
Satellite imagery
Semantic segmentation
Semantics
Online AccessGet full text

Cover

More Information
Summary:Semantic segmentation is a hot topic in computer vision and various deep learning networks are designed to achieve higher accuracy on that by fully exploring the capability of neural networks. This paper aims to address the issue and proposes the substructures with novelty for popular networks. Meanwhile, we present a cross‐channel structure, which simultaneously reduces parameter while the kernel size becomes larger. After that, to overcome the weakness of insufficient dataset which refers to satellite image data, we propose a feature attention mechanism with generative adversarial network to enhance the images' features. We show the recognition result on the satellite image dataset with a large picture. This paper evaluates substructures on the PASCAL VOC2012 dataset and improves the mIOU from 74.68% to 88.15%.
ISSN:1532-0626
1532-0634
DOI:10.1002/cpe.5920