A Fast and Compact 3-D CNN for Hyperspectral Image Classification

  title={A Fast and Compact 3-D CNN for Hyperspectral Image Classification},
  author={Muhammad Ahmad and A. Khan and Manuel Mazzara and Salvatore Distefano and Mohsin Ali and Muhammad Shahzad Sarfraz},
  journal={IEEE Geoscience and Remote Sensing Letters},
Hyperspectral images (HSIs) are used in a large number of real-world applications. HSI classification (HSIC) is a challenging task due to high interclass similarity, high intraclass variability, overlapping, and nested regions. The 2-D convolutional neural network (CNN) is a viable classification approach since HSIC depends on both spectral–spatial information. The 3-D CNN is a good alternative for improving the accuracy of HSIC, but it can be computationally intensive due to the volume and… 

Figures and Tables from this paper

SpectralNET: Exploring Spatial-Spectral WaveletCNN for Hyperspectral Image Classification

SpectralNET is proposed, a wavelet CNN, which is a variation of 2D CNN for multi-resolution HSI classification, and experiments performed with SpectralNET confirm the superiority of proposed Spectrals with respect to the state-of-the-art methods.

IRX-1D: A Simple Deep Learning Architecture for Remote Sensing Classifications

A simple deep learning architecture combining elements of Inception, ResNet and Xception networks is proposed, achieving high classification accuracy with limited training samples and comparison of classified image suggests different land cover classes are assigned to same area when compared with the classified image provided by the model trained using large training samples.

A Novel Stacked CNN for Malarial Parasite Detection in Thin Blood Smear Images

A novel Stacked Convolutional Neural Network architecture is proposed that improves the automatic detection of malaria without considering the hand-crafted features and is superior to the state-of-the-art models with 100% precision, 99.9% recall, and 99% f1-measure.

Fake News Stance Detection Using Deep Learning Architecture (CNN-LSTM)

A hybrid Neural Network architecture, that combines the capabilities of CNN and LSTM, is used with two different dimensionality reduction approaches, Principle Component Analysis (PCA) and Chi-Square, to determine the relative stance of a news article towards its headline.

3D residual spatial–spectral convolution network for hyperspectral remote sensing image classification

A hybrid 3D residual spatial–spectral convolution network (3D-RSSCN) is proposed to extract deep spatiospectral features using 3D CNN and ResNet18 architecture, and principal component analysis (PCA) is used as the preprocessing step for optimum spectral band extraction.

Hyperspectral Image Classification Based on Novel Hybridization of Spatial-Spectral-Superpixelwise Principal Component Analysis and Dense 2D-3D Convolutional Neural Network Fusion Architecture

Abstract We propose a hybridized technique named Spatial-Spectral-Superpixelwise PCA-based Dense 2D-3D CNN Fusion Architecture (3SPCA-D-2D-3D-CNN), that deals with the detailed and complex study of

A Full Tensor Network for Hyperspectral Image Classification

The proposed network can efficiently extract the high-level structural features directly from the original data and classify HSI in a single network without dimension reduction procedure in advance, and has a simple structure and holds fewer parameters than the state-of-the-art classification methods.

Spectral–Spatial Graph Attention Network for Semisupervised Hyperspectral Image Classification

The proposed SSGAT takes all samples as nodes and establishes an edge set among them to form a graph structure and computes attention coefficients between a node and its neighbor nodes and aggregates them to generate more discriminative features, thus improving the performance of HSI classification.

Knowledge and Geo-Object Based Graph Convolutional Network for Remote Sensing Semantic Segmentation

A knowledge and geo-object-based graph convolutional network (KGGCN) is proposed, which uses superpixel blocks as nodes of the graph network and combines prior knowledge with spatial correlations during information aggregation to overcome the distortion of the sample context.

Attention Mechanism Meets with Hybrid Dense Network for Hyperspectral Image Classification

The proposed solution aims at combining the core idea of 3D and 2D Inception net with the Attention mechanism to boost the HSIC CNN performance in a hybrid scenario and is able to selectively filter out the discriminative features critical for classification.



Spectral-Spatial Classification of Hyperspectral Imagery with 3D Convolutional Neural Network

A 3D convolutional neural network framework is proposed for accurate HSI classification, which is lighter, less likely to over-fit, and easier to train, and requires fewer parameters than other deep learning-based methods.

Hyperspectral Image Spatial Super-Resolution via 3D Full Convolutional Neural Network

3D convolution is used to exploit both the spatial context of neighboring pixels and spectral correlation of neighboring bands, such that spectral distortion when directly applying traditional CNN based SR algorithms to hyperspectral images in band-wise manners is alleviated.

An improved hybrid CNN for hyperspectral image classification

  • Yuting LiLin He
  • Environmental Science, Computer Science
    International Conference on Graphic and Image Processing
  • 2020
An improved hybrid CNN to enhance the classification performance, which involves global average pooling, skip connection and appropriate adjustments of the convolution kernels and overall structure is proposed.

HybridSN: Exploring 3-D–2-D CNN Feature Hierarchy for Hyperspectral Image Classification

A hybrid spectral CNN (HybridSN) for HSI classification is proposed that reduces the complexity of the model compared to the use of 3-D-CNN alone and is compared with the state-of-the-art hand-crafted as well as end-to-end deep learning-based methods.

HSI-CNN: A Novel Convolution Neural Network for Hyperspectral Image

A novel convolutional neural network framework for the characteristics of hyperspectral image data called HSI-CNN, which can also provides ideas for the processing of one-dimensional data.

Multi-scale 3D deep convolutional neural network for hyperspectral image classification

  • Mingyi HeBo LiHuahui Chen
  • Computer Science, Environmental Science
    2017 IEEE International Conference on Image Processing (ICIP)
  • 2017
A Multiscale 3D deep convolutional neural network (M3D-DCNN) is proposed for HSI classification, which could jointly learn both 2D Multi-scale spatial feature and 1D spectral feature from HSI data in an end-to-end approach, promising to achieve better results with large-scale dataset.

Dual-Path Siamese CNN for Hyperspectral Image Classification With Limited Training Samples

  • Lin HuangYushi Chen
  • Environmental Science, Computer Science
    IEEE Geoscience and Remote Sensing Letters
  • 2021
The proposed classification framework is a combination of extended morphological profiles, CNN, siamese network, and spectral–spatial feature fusion and results reveal that the proposed methods provide the competitive results in terms of classification accuracy, especially with limited training samples.

3-D Deep Learning Approach for Remote Sensing Image Classification

The aim of this paper is first to explore the performance of DL architectures for the RS hyperspectral data set classification and second to introduce a new 3-D DL approach that enables a joint spectral and spatial information process.

A semi-supervised convolutional neural network for hyperspectral image classification

A novel semi-supervised convolutional neural network is proposed for the classification of hyperspectral image that can automatically learn features from complex hyperspectRAL image data structures and simultaneously minimize the sum of supervised and unsupervised cost functions.

Three-Dimensional Gabor Wavelets for Pixel-Based Hyperspectral Imagery Classification

  • L. ShenSen Jia
  • Environmental Science
    IEEE Transactions on Geoscience and Remote Sensing
  • 2011
A 3-D Gabor-wavelet-based approach for pixel-based hyperspectral imagery classification and results show that the method achieves as high as 96.04% and 95.36% accuracies, respectively, even when only few samples per class are labeled.