3D atrous spatial pyramid pooling based multi-scale feature fusion network for hyperspectral image classification

Tianxing Zhu; Qin Liu; Lixiang Zhang

doi:10.1117/12.3010238

15 November 2023 3D atrous spatial pyramid pooling based multi-scale feature fusion network for hyperspectral image classification

Tianxing Zhu, Qin Liu, Lixiang Zhang

Author Affiliations +

Proceedings Volume 12815, International Conference on Remote Sensing, Mapping, and Geographic Systems (RSMG 2023); 128150X (2023) https://doi.org/10.1117/12.3010238
Event: International Conference on Remote Sensing, Mapping, and Geographic Systems (RSMG 2023), 2023, Kaifeng, China

Abstract

The convolutional neural network (CNN) has been widely adopted in the processing of hyperspectral images (HSI) due to its remarkable capabilities in feature extraction. However, CNN-based methods for HSI classification still face the challenge of extracting spectral-spatial features that are more effective while considering all spectral bands. This paper reviews various spectral and spectral-spatial classifiers that have been developed to overcome the challenges of HSI classification, including the limitations of handcrafted features and the problems caused by spectral variability and nonlinearity. The paper proposes a novel method, the 3D ASPP Multi-Scale Feature Fusion Network (3A-MFFN), which simultaneously preserves high-resolution spatial detail information and high-level semantic information of HSI. The proposed network extracts and fuses multi-scale features for HSI classification using a multi-scale 3D ASPP convolution block to fuse different levels of feature output. The paper presents experiments and analysis to demonstrate the effectiveness of the proposed method. The results suggest that the 3A-MFFN algorithm can significantly enhance the classification accuracy of HSI with limited samples, making it a valuable contribution to the hyperspectral remote sensing community.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Tianxing Zhu, Qin Liu, and Lixiang Zhang "3D atrous spatial pyramid pooling based multi-scale feature fusion network for hyperspectral image classification", Proc. SPIE 12815, International Conference on Remote Sensing, Mapping, and Geographic Systems (RSMG 2023), 128150X (15 November 2023); https://doi.org/10.1117/12.3010238

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
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Convolution

3D modeling

Feature extraction

Hyperspectral imaging

Feature fusion

Image classification

Image fusion

Show All Keywords

Keywords/Phrases

Search In:

Publication Years