A Novel 2D-3D CNN with Spectral-Spatial Multi-Scale Feature Fusion for Hyperspectral Image Classification

Multifarious hyperspectral image (HSI) classification methods based on convolutional neural networks (CNN) have been gradually proposed and achieve a promising classification performance. However, hyperspectral image classification still suffers from various challenges, including abundant redundant...

Full description

Saved in:
Bibliographic Details
Published in:Remote sensing (Basel, Switzerland) Switzerland), 2021-11, Vol.13 (22), p.4621
Main Authors: Liu, Dongxu, Han, Guangliang, Liu, Peixun, Yang, Hang, Sun, Xinglong, Li, Qingqing, Wu, Jiajia
Format: Article
Language:English
Subjects:
2D-3D CNN
Accuracy
Artificial neural networks
attention module
Classification
Classification schemes
Design
Feature extraction
Feature selection
hyperspectral image classification
Hyperspectral imaging
Image classification
Image reconstruction
Modules
multi-level features
multi-scale features
Neural networks
Remote sensing
Representations
Semantics
Spatial distribution
Spectra
Streams
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Multifarious hyperspectral image (HSI) classification methods based on convolutional neural networks (CNN) have been gradually proposed and achieve a promising classification performance. However, hyperspectral image classification still suffers from various challenges, including abundant redundant information, insufficient spectral-spatial representation, irregular class distribution, and so forth. To address these issues, we propose a novel 2D-3D CNN with spectral-spatial multi-scale feature fusion for hyperspectral image classification, which consists of two feature extraction streams, a feature fusion module as well as a classification scheme. First, we employ two diverse backbone modules for feature representation, that is, the spectral feature and the spatial feature extraction streams. The former utilizes a hierarchical feature extraction module to capture multi-scale spectral features, while the latter extracts multi-stage spatial features by introducing a multi-level fusion structure. With these network units, the category attribute information of HSI can be fully excavated. Then, to output more complete and robust information for classification, a multi-scale spectral-spatial-semantic feature fusion module is presented based on a Decomposition-Reconstruction structure. Last of all, we innovate a classification scheme to lift the classification accuracy. Experimental results on three public datasets demonstrate that the proposed method outperforms the state-of-the-art methods.
ISSN:2072-4292
2072-4292
DOI:10.3390/rs13224621