3-D Deep Learning Approach for Remote Sensing Image Classification

By A. Ben Hamida, Alexandre Benoit, Patrick Lambert, C. Ben Amar. IEEE Transactions on Geoscience and Remote Sensing, 2018

765 citations recorded by the source corpus.

Recently, a variety of approaches have been enriching the field of remote sensing (RS) image processing and analysis. Unfortunately, existing methods remain limited to the rich spatiospectral content of today’s large data sets. It would seem intriguing to resort to deep learning (DL)-based approaches at this stage with regard to their ability to offer accurate semantic interpretation of the data. However, the specificity introduced by the coexistence of spectral and spatial content in the RS data sets widens the scope of the challenges presented to adapt DL methods to these contexts. Therefore, 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 set of 3-D schemes is proposed and evaluated. Experimental results based on well-known hyperspectral data sets demonstrate that the proposed method is able to achieve a better classification rate than state-of-the-art methods with lower computational costs.

Subfields: Ecology, Media Technology

View Paper: https://www.semanticscholar.org/paper/f748311d212786bfe6b6c6878d633db04bbe4e9a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology

In Media Technology

Citations in the corpus

Spectral–Spatial Morphological Attention Transformer for Hyperspectral Image Classification (2023, methodology): “D CNNs (CNN3Ds) [31] have demonstrated success in the classification of HSI data.”
Multimodal Fusion Transformer for Remote Sensing Image Classification (2022, methodology): “One-dimensional convolution (CNN1D) [33], two-dimensional convolution (CNN2D) [34], and three-dimensional convolution (CNN3D) [35] have shown success in HSI and LiDAR data classiﬁcation.”
Hyperspectral Image Classification—Traditional to Deep Models: A Survey for Future Prospects (2021, methodology): “TABLE III: Classification results obtained by RF [64], MLR [308], SVM [309], MLP [310], RNN [59], LSTM [311], GRU [312], CNN-1D [313], CNN-2D [314], CNN-3D [315], HybridSN [200], and MorphCNN [183] on the disjoint train-test dataset for the PU scene.”
Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series (2018, background): “The main successful application of CNNs in remote sensing remains the classification of hyperspectral images, where 2D-CNNs across the spatial dimension have also been tested [31], as well as 1D-CNNs across the spectral dimension [32], and even 3D-CNNs across spectral and spatial dimensions [33,34].”
Deep Learning for Classification of Hyperspectral Data: A Comparative Review (2019, methodology): “◗ 2D + 1D CNN [63]: It required no modification.”
Attention-Based Adaptive Spectral–Spatial Kernel ResNet for Hyperspectral Image Classification (2020, background): “[29] presented a 3-D deep learning architecture, which takes a 3-D volume as input to extract joint spatial–spectral features.”
HybridSN: Exploring 3-D–2-D CNN Feature Hierarchy for Hyperspectral Image Classification (2019, methodology): “The results of the proposed H ybrid SN model are compared with the most widely used supervised methods, such as SVM [33], 2-D-CNN [34], 3-D-CNN [35], multi-scale 3-D deep convolutional neural network (M3D)-CNN [36], and SSRN [27].”
A Spectral-Spatial-Dependent Global Learning Framework for Insufficient and Imbalanced Hyperspectral Image Classification (2021, methodology): “Many supervised classification methods based on deep learning are used to extract the spectral and spatial features, such as 3D-CNN and multiscale CNNs [17], [18].”