Three-dimensional densely connected convolutional network for hyperspectral remote sensing image classification

Zhang, Chunju; Li, Guandong; Du, Shihong; Tan, Wuzhou; Gao, Fei

doi:10.1117/1.jrs.13.016519

Cited by 48 publications

(15 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to test the impact of the number of training samples on models, just like the experiment of the effect with different ratios of training samples in Reference [37], we also divided the datasets into different proportions (2 : 1 : 7, 3 : 1 : 6, 4 : 1 : 5, 5 : 1 : 4). To obtain the best results, the epochs of different ratios are {100, 80, 60, 60}, respectively.…”

Section: Effect Of Different Ratios Of Training Validation and Test mentioning

confidence: 99%

“…It fused the output of different hierarchical layers to improve the classification accuracy. According to dense convolutional network in References [36], Reference [37] proposed the three-dimensional densely connected convolutional network (3D-DenseNet). Through the densely connected structure, the 3D-DenseNet was deeper in structure and could learn more robust spectral-spatial features.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Three-Dimensional ResNeXt Network Using Feature Fusion and Label Smoothing for Hyperspectral Image Classification

Cui

Gan

et al. 2020

Sensors

View full text Add to dashboard Cite

In recent years, deep learning methods have been widely used in the hyperspectral image (HSI) classification tasks. Among them, spectral-spatial combined methods based on the three-dimensional (3-D) convolution have shown good performance. However, because of the three-dimensional convolution, increasing network depth will result in a dramatic rise in the number of parameters. In addition, the previous methods do not make full use of spectral information. They mostly use the data after dimensionality reduction directly as the input of networks, which result in poor classification ability in some categories with small numbers of samples. To address the above two issues, in this paper, we designed an end-to-end 3D-ResNeXt network which adopts feature fusion and label smoothing strategy further. On the one hand, the residual connections and split-transform-merge strategy can alleviate the declining-accuracy phenomenon and decrease the number of parameters. We can adjust the hyperparameter cardinality instead of the network depth to extract more discriminative features of HSIs and improve the classification accuracy. On the other hand, in order to improve the classification accuracies of classes with small numbers of samples, we enrich the input of the 3D-ResNeXt spectral-spatial feature learning network by additional spectral feature learning, and finally use a loss function modified by label smoothing strategy to solve the imbalance of classes. The experimental results on three popular HSI datasets demonstrate the superiority of our proposed network and an effective improvement in the accuracies especially for the classes with small numbers of training samples.

show abstract

Section: Effect Of Different Ratios Of Training Validation and Test mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Three-Dimensional ResNeXt Network Using Feature Fusion and Label Smoothing for Hyperspectral Image Classification

Cui

Gan

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Inspired by this, many recent HSI classification methods have also begun to take advantage of residual learning for training deeper neural networks. For example, Zhang et al [25] employed a 3D densely connected convolutional neural network (3D-DenseNet) to learn the spectral-spatial features of HSIs for classification. With its densely connected structure, the deeper network can be easier to train.…”

Section: Introductionmentioning

confidence: 99%

Boosting Few-Shot Hyperspectral Image Classification Using Pseudo-Label Learning

et al. 2021

View full text Add to dashboard Cite

Deep neural networks have underpinned much of the recent progress in the field of hyperspectral image (HSI) classification owing to their powerful ability to learn discriminative features. However, training a deep neural network often requires the availability of a large number of labeled samples to mitigate over-fitting, and these labeled samples are not always available in practical applications. To adapt the deep neural network-based HSI classification approach to cases in which only a very limited number of labeled samples (i.e., few or even only one labeled sample) are provided, we propose a novel few-shot deep learning framework for HSI classification. In order to mitigate over-fitting, the framework borrows supervision from an auxiliary set of unlabeled samples with soft pseudo-labels to assist the training of the feature extractor on few labeled samples. By considering each labeled sample as a reference agent, the soft pseudo-label is assigned by computing the distances between the unlabeled sample and all agents. To demonstrate the effectiveness of the proposed method, we evaluate it on three benchmark HSI classification datasets. The results indicate that our method achieves better performance relative to existing competitors in few-shot and one-shot settings.

show abstract

“…(1) In the process of supervised learning, the imbalance between high-dimensional data and limited training samples can easily lead to the phenomenon that classification results decline with the increase of dimensions, which is called the curse of dimensionality [8]. (2) The high cost of manual labeling of HSIs leads to the shortage of label samples [9]. (3) The spatial layout of HSIs is complicated.…”

Section: Introductionmentioning

confidence: 99%

“…Zhong et al proposed the methods based on 3-D residual connections for the classification of HSIs [57]. Zhang et al introduced the 3-D densely connected convolutional network to extract spectral-spatial feature of HSIs [9]. Wang et al proposed a deep and fast 3-D CNN framework based on dense connectivity, and obtained satisfactory results [58].…”

Section: Introductionmentioning

confidence: 99%

SSDANet: Spectral-Spatial Three-Dimensional Convolutional Neural Network for Hyperspectral Image Classification

et al. 2020

View full text Add to dashboard Cite

Recently, the classification of hyperspectral images has made great process. Especially, the classification methods based on three-dimensional convolutional neural network have remarkable performance due to the uniqueness of hyperspectral images. However, the hyperspectral classification still faces great challenges due to a series of problems such as the insufficient extraction of spectral-spatial features, the lack of labeled samples, the large amount of noise, the tendency of overfitting and so on. Therefore, SSDANet is proposed to solve the above problems and promote the further development of hyperspectral classification technology based on deep learning. SSDANet is a spectral-spatial three-dimensional convolutional neural network with a deep and wide structure that can significantly improve classification performance. In SSDANet, the spectral-spatial dense connectivity is put forward to protect the integrity of information. It is made up of the spectral branch and the spatial branch, which can learn and reuse the spectralspatial features. Besides, the spectral-spatial attention mechanism is proposed to adapt the special structure of hyperspectral images. It can excite important spectral-spatial information and suppress unimportant spectral-spatial information. In addition, a series of optimization methods including data augmentation, batch normalization, dropout, exponential decay learning rate, and L2 regularization are adopted to alleviate the problem of overfitting and improve the classification results. To verify the performance of SSDANet, experiments were implemented on two widely used datasets-Pavia University and Indian Pines. Under the condition of limited labeled samples, the classification evaluation indexes of OA, AA, and Kappa on the two datasets all exceeded 99%, reaching state-of-the-art performance. INDEX TERMS artificial intelligence, hyperspectral imaging, image processing, pattern recognition, remote sensing

show abstract

Three-dimensional densely connected convolutional network for hyperspectral remote sensing image classification

Cited by 48 publications

References 22 publications

Three-Dimensional ResNeXt Network Using Feature Fusion and Label Smoothing for Hyperspectral Image Classification

Three-Dimensional ResNeXt Network Using Feature Fusion and Label Smoothing for Hyperspectral Image Classification

Boosting Few-Shot Hyperspectral Image Classification Using Pseudo-Label Learning

SSDANet: Spectral-Spatial Three-Dimensional Convolutional Neural Network for Hyperspectral Image Classification

Contact Info

Product

Resources

About