Spectral Segmentation Multi-Scale Feature Extraction Residual Networks for Hyperspectral Image Classification

Hyperspectral image (HSI) classification is a vital task in hyperspectral image processing and applications. Convolutional neural networks (CNN) are becoming an effective approach for categorizing hyperspectral remote sensing images as deep learning technology advances. However, traditional CNN usua...

Full description

Saved in:
Bibliographic Details
Published in:Remote sensing (Basel, Switzerland) Switzerland), 2023-09, Vol.15 (17), p.4219
Main Authors: Wang, Jiamei, Ren, Jiansi, Peng, Yinbin, Shi, Meilin
Format: Article
Language:English
Subjects:
Accuracy
Analysis
Artificial neural networks
Band spectra
Classification
Deep learning
Feature extraction
grouped convolution
hyperspectral image (HSI)
Hyperspectral imaging
Image classification
Image processing
Image segmentation
Information processing
Information retrieval
Machine learning
Modules
multi-scale
Neural networks
Receptive field
Remote sensing
residual structure
Spatial data
Spectral bands
spectral segmentation
Support vector machines
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:Hyperspectral image (HSI) classification is a vital task in hyperspectral image processing and applications. Convolutional neural networks (CNN) are becoming an effective approach for categorizing hyperspectral remote sensing images as deep learning technology advances. However, traditional CNN usually uses a fixed kernel size, which limits the model’s capacity to acquire new features and affects the classification accuracy. Based on this, we developed a spectral segmentation-based multi-scale spatial feature extraction residual network (MFERN) for hyperspectral image classification. MFERN divides the input data into many non-overlapping sub-bands by spectral bands, extracts features in parallel using the multi-scale spatial feature extraction module MSFE, and adds global branches on top of this to obtain global information of the full spectral band of the image. Finally, the extracted features are fused and sent into the classifier. Our MSFE module has multiple branches with increasing ranges of the receptive field (RF), enabling multi-scale spatial information extraction at both fine- and coarse-grained levels. On the Indian Pines (IP), Salinas (SA), and Pavia University (PU) HSI datasets, we conducted extensive experiments. The experimental results show that our model has the best performance and robustness, and our proposed MFERN significantly outperforms other models in terms of classification accuracy, even with a small amount of training data.
ISSN:2072-4292
2072-4292
DOI:10.3390/rs15174219