Fusion attention-ASPP and context semantic for complex scene semantic segmentation

YingHao Lin; ShiHao Zhao; Kun Cai; BaoJun Qiao

doi:10.1117/12.2674281

28 April 2023 Fusion attention-ASPP and context semantic for complex scene semantic segmentation

YingHao Lin, ShiHao Zhao, Kun Cai, BaoJun Qiao

Author Affiliations +

Proceedings Volume 12626, International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022); 126260V (2023) https://doi.org/10.1117/12.2674281
Event: International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022), 2022, Zhuhai, China

Abstract

Semantic segmentation is widely used in remote sensing data extraction and classification. Existing semantic segmentation networks focus on capturing contextual information in many different ways, simply fusing features at different levels, and ultimately improving the accuracy of semantic segmentation. However, low-level semantic features lack spatial context guidance, and high-level semantic features tend to encode large objects with coarse spatial details, making segmentation results prone to losing fine details. In this paper, we analyze the advantages and disadvantages of different levels of feature maps, and enhance the feature representation from two aspects to solve this problem. On the one hand, inspired by the architectural idea of atrous spatial pyramic pooling (ASPP), we adjust the structure of ASPP module and add the attention module to ASPP, and a new Attention-ASPP(AASPP) module is constructed in this paper. On the other hand, feature information such as boundary contours is enhanced by channel attention modeling, thereby improving local detail representation. Comprehensive experimental results show that our model framework achieves excellent segmentation performance on two public datasets, WHU building dataset and ISPRS Potsdam dataset.

Citation Download Citation

YingHao Lin, ShiHao Zhao, Kun Cai, and BaoJun Qiao "Fusion attention-ASPP and context semantic for complex scene semantic segmentation", Proc. SPIE 12626, International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022), 126260V (28 April 2023); https://doi.org/10.1117/12.2674281

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Semantics

Image segmentation

Convolution

Education and training

Remote sensing

Data modeling

Convolutional neural networks

Show All Keywords

Keywords/Phrases

Search In:

Publication Years