Multitask Deep Learning With Spectral Knowledge for Hyperspectral Image Classification

@article{Liu2020MultitaskDL,
  title={Multitask Deep Learning With Spectral Knowledge for Hyperspectral Image Classification},
  author={Shengjie Liu and Qian Shi},
  journal={IEEE Geoscience and Remote Sensing Letters},
  year={2020},
  volume={17},
  pages={2110-2114}
}
  • Shengjie Liu, Q. Shi
  • Published 2020
  • Computer Science
  • IEEE Geoscience and Remote Sensing Letters
In this letter, we propose a multitask deep learning method for the classification of multiple hyperspectral data in a single training. Deep learning models have achieved promising results on hyperspectral image classification, but their performance highly relies on sufficient labeled samples that are scarce on hyperspectral images. However, samples from multiple data sets might be sufficient to train one deep learning model, thereby improving its performance. To do so, we trained an identical… Expand
Comparative analysis of deep transfer learning performance on crop classification
TLDR
A comparative analysis of various transfer learning strategies for the domain of crop classification using three well-known neural networks and evaluating various strategies such as training the model from scratch using random weight initialization, and simply using the pretrained models as feature extractors. Expand
Detection and Classification of Early Decay on Blueberry Based on Improved Deep Residual 3D Convolutional Neural Network in Hyperspectral Images
TLDR
A novel strategy to enhance the hyperspectral image sample data, which can improve the training effect and reduce the number of parameters by half and the training time by about 10% is proposed. Expand
Rule-Based Classification of Hyperspectral Imaging Data
TLDR
A general classification approach based on the shape of spectral signatures is presented, in contrast to classical classification approaches, where not only reflectance values are considered, but also parameters such as curvature points, curvature values, and the curvature behavior of spectral signature are used. Expand
Stack Attention-Pruning Aggregates Multiscale Graph Convolution Networks for Hyperspectral Remote Sensing Image Classification
TLDR
A stack attention-pruning multiscale aggregates graph convolution framework (SAP-MAGACN) that can effectively disentangle the complex space structure of remote sensing images and capture the rich structural semantics, and gradually produce the discriminative embedded features and effectively distinguish the categories of boundary pixels. Expand
Few-Shot Hyperspectral Image Classification With Unknown Classes Using Multitask Deep Learning
TLDR
A multitask deep learning method that simultaneously conducts classification and reconstruction in the open world (named MDL4OW) where unknown classes may exist, and achieves more accurate hyperspectral image classification, especially under the few-shot context. Expand
Transfer Vision Patterns for Multi-Task Pixel Learning
  • Xiaoya Zhang, Ling Zhou, Yong Li, Zhen Cui, Jin Xie, Jian Yang
  • 2021
Multi-task pixel perception is one of the most important topics in the field of machine intelligence. Inspired by the observation of cross-task interdependencies of visual patterns, we propose aExpand

References

SHOWING 1-10 OF 28 REFERENCES
Deep Learning-Based Classification of Hyperspectral Data
TLDR
The concept of deep learning is introduced into hyperspectral data classification for the first time, and a new way of classifying with spatial-dominated information is proposed, which is a hybrid of principle component analysis (PCA), deep learning architecture, and logistic regression. Expand
MugNet: Deep learning for hyperspectral image classification using limited samples
TLDR
A small-scale data based method, multi-grained network (MugNet), to explore the application of deep learning approaches in hyperspectral image classification and is built upon the basis of a very simple network which does not include many hyperparameters for tuning. Expand
Active Learning With Convolutional Neural Networks for Hyperspectral Image Classification Using a New Bayesian Approach
TLDR
A new AL-guided classification model is developed that exploits both the spectral information and the spatial-contextual information in the hyperspectral data that makes use of recently developed Bayesian CNNs. Expand
Learning and Transferring Deep Joint Spectral–Spatial Features for Hyperspectral Classification
TLDR
Experiments demonstrate that the learned deep joint spectral–spatial features are discriminative, and competitive classification results can be achieved when compared with state-of-the-art methods. Expand
Self-Taught Feature Learning for Hyperspectral Image Classification
TLDR
This paper studied two self-taught learning frameworks for HSI classification, one of which is a shallow approach that uses independent component analysis and the second is a three-layer stacked convolutional autoencoder. Expand
Going Deeper With Contextual CNN for Hyperspectral Image Classification
TLDR
A novel deep convolutional neural network that is deeper and wider than other existing deep networks for hyperspectral image classification, called contextual deep CNN, can optimally explore local contextual interactions by jointly exploiting local spatio-spectral relationships of neighboring individual pixel vectors. Expand
Cascaded Recurrent Neural Networks for Hyperspectral Image Classification
TLDR
This paper proposes a cascaded RNN model using gated recurrent units to explore the redundant and complementary information of HSIs and extends the proposed model to its spectral–spatial counterpart by incorporating some convolutional layers. Expand
Spectral–Spatial Residual Network for Hyperspectral Image Classification: A 3-D Deep Learning Framework
TLDR
An end-to-end spectral–spatial residual network that takes raw 3-D cubes as input data without feature engineering for hyperspectral image classification and achieves the state-of-the-art HSI classification accuracy in agricultural, rural–urban, and urban data sets. Expand
A new deep convolutional neural network for fast hyperspectral image classification
TLDR
A new CNN architecture for the classification of hyperspectral images is presented, a 3-D network that uses both spectral and spatial information and implements a border mirroring strategy to effectively process border areas in the image. Expand
Spectral Classification of a Set of Hyperspectral Images using the Convolutional Neural Network, in a Single Training
TLDR
A new Fast Spectral classification algorithm based on CNN is proposed which allows to build a composite image from multiple hyperspectral images, then trains the model only once on the composite image. Expand
...
1
2
3
...