2SRS: Two-Stream Residual Separable Convolution Neural Network for Hyperspectral Image Classification

Zharfan Zahisham, Kian Ming Lim*, Voon Chet Koo, Yee Kit Chan, Chin Poo Lee

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

8 Citations (Scopus)

Abstract

Typically, hyperspectral image suffers from redundant information, data scarcity, and class imbalance problems. This letter proposes a hyperspectral image classification framework named a two-stream residual separable convolution (2SRS) network that aims to mitigate these problems. Principal component analysis (PCA) is first employed to reduce the spectral dimension of the hyperspectral image. Subsequently, the data scarcity and class imbalance problems are overcome via spatial and spectral data augmentations. A novel spectral data creation from image patches is proposed. The augmented samples are fed into the proposed 2SRS network for hyperspectral image classification. We evaluated the proposed method on three benchmark datasets, namely, 1) Indian Pines (IP); 2) Pavia University; and 3) Salinas Scene (SA). The proposed method achieved state-of-the-art performance in terms of overall accuracy (OA), average accuracy (AA), and Kappa coefficient (Kappa) for both 30% and 10% training set ratios.

Original languageEnglish
Article number5501505
JournalIEEE Geoscience and Remote Sensing Letters
Volume20
DOIs
Publication statusPublished - 2023
Externally publishedYes

Keywords

  • Convolutional neural networks (CNNs)
  • hyperspectral image classification
  • remote sensing
  • residual learning
  • separable 2-D-CNN

Fingerprint

Dive into the research topics of '2SRS: Two-Stream Residual Separable Convolution Neural Network for Hyperspectral Image Classification'. Together they form a unique fingerprint.

Cite this