Wavelet-based extended morphological profile and deep autoencoder for hyperspectral image classification

In this paper, we propose a novel scheme to learn high-level representative features and conduct classification for hyperspectral image (HSI) data in an automatic fashion. The proposed method is a collaboration of a wavelet-based extended morphological profile (WTEMP) and a deep autoencoder (DAE) (“WTEMP-DAE”), with the aim of exploiting the discriminative capability of DAE when using WTEMP features as the input. Each part of WTEMP-DAE is ingenious and contributes to the final classification performance. Specifically, in WTEMP-DAE, the spatial information is extracted from the WTEMP, which is then joined with the wavelet denoised spectral information to form the spectral-spatial description of HSI data. The obtained features are fed into DAE as the original input, where the good weights and bias of the network are initialized through unsupervised pre-training. Once the pre-training is completed, the reconstruction layers are discarded and a logistic regression (LR) layer is added to the top of the network to perform supervised fine-tuning and classification. Experimental results on two real HSI data sets demonstrate that the proposed strategy improves classification performance in comparison with other state-of-the-art hand-crafted feature extractors and their combinations.

Download Full-text

A DIVERSIFIED DEEP BELIEF NETWORK FOR HYPERSPECTRAL IMAGE CLASSIFICATION

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xli-b7-443-2016 ◽

2016 ◽

Vol XLI-B7 ◽

pp. 443-449

Author(s):

P. Zhong ◽

Z. Q. Gong ◽

C. Schönlieb

Keyword(s):

Image Classification ◽

Real World ◽

Hyperspectral Image ◽

Classification Performance ◽

Fine Tuning ◽

Hyperspectral Image Classification ◽

Latent Factors ◽

Deep Model ◽

Training Samples ◽

Limited Training Samples

In recent years, researches in remote sensing demonstrated that deep architectures with multiple layers can potentially extract abstract and invariant features for better hyperspectral image classification. Since the usual real-world hyperspectral image classification task cannot provide enough training samples for a supervised deep model, such as convolutional neural networks (CNNs), this work turns to investigate the deep belief networks (DBNs), which allow unsupervised training. The DBN trained over limited training samples usually has many “dead” (never responding) or “potential over-tolerant” (always responding) latent factors (neurons), which decrease the DBN’s description ability and thus finally decrease the hyperspectral image classification performance. This work proposes a new diversified DBN through introducing a diversity promoting prior over the latent factors during the DBN pre-training and fine-tuning procedures. The diversity promoting prior in the training procedures will encourage the latent factors to be uncorrelated, such that each latent factor focuses on modelling unique information, and all factors will be summed up to capture a large proportion of information and thus increase description ability and classification performance of the diversified DBNs. The proposed method was evaluated over the well-known real-world hyperspectral image dataset. The experiments demonstrate that the diversified DBNs can obtain much better results than original DBNs and comparable or even better performances compared with other recent hyperspectral image classification methods.

Download Full-text

Hyperspectral Image Classification Based on Sparse Superpixel Graph

Remote Sensing ◽

10.3390/rs13183592 ◽

2021 ◽

Vol 13 (18) ◽

pp. 3592

Author(s):

Yifei Zhao ◽

Fengqin Yan

Keyword(s):

Classification Accuracy ◽

Spatial Information ◽

Hyperspectral Image ◽

Hyperspectral Images ◽

Spectral Information ◽

Hyperspectral Image Classification ◽

Computational Burden ◽

Large Size ◽

High Classification Accuracy ◽

Good Classification

Hyperspectral image (HSI) classification is one of the major problems in the field of remote sensing. Particularly, graph-based HSI classification is a promising topic and has received increasing attention in recent years. However, graphs with pixels as nodes generate large size graphs, thus increasing the computational burden. Moreover, satisfactory classification results are often not obtained without considering spatial information in constructing graph. To address these issues, this study proposes an efficient and effective semi-supervised spectral-spatial HSI classification method based on sparse superpixel graph (SSG). In the constructed sparse superpixels graph, each vertex represents a superpixel instead of a pixel, which greatly reduces the size of graph. Meanwhile, both spectral information and spatial structure are considered by using superpixel, local spatial connection and global spectral connection. To verify the effectiveness of the proposed method, three real hyperspectral images, Indian Pines, Pavia University and Salinas, are chosen to test the performance of our proposal. Experimental results show that the proposed method has good classification completion on the three benchmarks. Compared with several competitive superpixel-based HSI classification approaches, the method has the advantages of high classification accuracy (>97.85%) and rapid implementation (<10 s). This clearly favors the application of the proposed method in practice.

Download Full-text

Active Semi-Supervised Random Forest for Hyperspectral Image Classification

Remote Sensing ◽

10.3390/rs11242974 ◽

2019 ◽

Vol 11 (24) ◽

pp. 2974 ◽

Cited By ~ 3

Author(s):

Youqiang Zhang ◽

Guo Cao ◽

Xuesong Li ◽

Bisheng Wang ◽

Peng Fu

Keyword(s):

Random Forest ◽

Spatial Information ◽

Hyperspectral Image ◽

Hyperspectral Data ◽

Data Sets ◽

Great Success ◽

Full Potential ◽

Hyperspectral Image Classification ◽

Unified Framework ◽

Proposed Model

Random forest (RF) has obtained great success in hyperspectral image (HSI) classification. However, RF cannot leverage its full potential in the case of limited labeled samples. To address this issue, we propose a unified framework that embeds active learning (AL) and semi-supervised learning (SSL) into RF (ASSRF). Our aim is to utilize AL and SSL simultaneously to improve the performance of RF. The objective of the proposed method is to use a small number of manually labeled samples to train classifiers with relative high classification accuracy. To achieve this goal, a new query function is designed to query the most informative samples for manual labeling, and a new pseudolabeling strategy is introduced to select some samples for pseudolabeling. Compared with other AL- and SSL-based methods, the proposed method has several advantages. First, ASSRF utilizes the spatial information to construct a query function for AL, which can select more informative samples. Second, in addition to providing more labeled samples for SSL, the proposed pseudolabeling method avoids bias caused by AL-labeled samples. Finally, the proposed model retains the advantages of RF. To demonstrate the effectiveness of ASSRF, we conducted experiments on three real hyperspectral data sets. The experimental results have shown that our proposed method outperforms other state-of-the-art methods.

Download Full-text

Variational Generative Adversarial Network with Crossed Spatial and Spectral Interactions for Hyperspectral Image Classification

Remote Sensing ◽

10.3390/rs13163131 ◽

2021 ◽

Vol 13 (16) ◽

pp. 3131

Author(s):

Zhongwei Li ◽

Xue Zhu ◽

Ziqi Xin ◽

Fangming Guo ◽

Xingshuai Cui ◽

...

Keyword(s):

Image Classification ◽

Spatial Information ◽

Hyperspectral Image ◽

Classification Performance ◽

Generative Adversarial Networks ◽

Hyperspectral Image Classification ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Spatial Features ◽

Virtual Samples

Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) have been widely used in hyperspectral image classification (HSIC) tasks. However, the generated HSI virtual samples by VAEs are often ambiguous, and GANs are prone to the mode collapse, which lead the poor generalization abilities ultimately. Moreover, most of these models only consider the extraction of spectral or spatial features. They fail to combine the two branches interactively and ignore the correlation between them. Consequently, the variational generative adversarial network with crossed spatial and spectral interactions (CSSVGAN) was proposed in this paper, which includes a dual-branch variational Encoder to map spectral and spatial information to different latent spaces, a crossed interactive Generator to improve the quality of generated virtual samples, and a Discriminator stuck with a classifier to enhance the classification performance. Combining these three subnetworks, the proposed CSSVGAN achieves excellent classification by ensuring the diversity and interacting spectral and spatial features in a crossed manner. The superior experimental results on three datasets verify the effectiveness of this method.

Download Full-text

Hyperspectral image classification using wavelet transform-based smooth ordering

International Journal of Wavelets Multiresolution and Information Processing ◽

10.1142/s0219691319500504 ◽

2019 ◽

Vol 17 (06) ◽

pp. 1950050 ◽

Cited By ~ 6

Author(s):

Lina Yang ◽

Hailong Su ◽

Cheng Zhong ◽

Zuqiang Meng ◽

Huiwu Luo ◽

...

Keyword(s):

Wavelet Transform ◽

Spatial Information ◽

Hyperspectral Image ◽

Interpolation Method ◽

Computational Cost ◽

Classification Performance ◽

Hyperspectral Image Classification ◽

Measurement Results ◽

Improve Accuracy ◽

Main Components

To efficiently improve the accuracy of hyperspectral image (HSI) classification, the spatial information is usually fused with spectral information so that the classification performance can be enhanced. In this paper, we propose a new classification method called wavelet transform-based smooth ordering (WTSO). WTSO consists of three main components: wavelet transform for feature extraction, spectral–spatial based similarity measurement, smooth ordering based 1D embedding, and construction of final classifier using interpolation scheme. Specifically, wavelet transform is first imposed to decompose the HSI signal into approximate coefficients (ACs) and details coefficients (DCs). Then, to measure the similar level of pairwise samples, a novel metric is defined on the ACs, where the spatial information serves as the prior knowledge. Next, according to the measurement results, smooth ordering is applied so that the samples are aligned in a 1D space (called 1D embedding). Finally, since the reordering samples are smooth, the labels of test samples can be recovered using the simple 1D interpolation method. In the last step, in order to reduce the bias and improve accuracy, the final classifier is constructed using multiple 1D embeddings. The use of wavelet transform in WTSO can also reduce the high dimensionality of HSI data. By converting the hight-dimensional samples into a 1D ordering sequence, WTSO can reduce the computational cost, and simultaneously perform classification for the test samples. Note that in WTSO, the smooth ordering based 1D embedding and interpolation are executed in an iterative manner. And they will be terminated after finite steps. The proposed method is experimentally demonstrated on two real HSI datasets: IndianPines and University of Pavia, achieving promising results.

Download Full-text

Spatial–Spectral Squeeze-and-Excitation Residual Network for Hyperspectral Image Classification

Remote Sensing ◽

10.3390/rs11070884 ◽

2019 ◽

Vol 11 (7) ◽

pp. 884 ◽

Cited By ~ 13

Author(s):

Li Wang ◽

Jiangtao Peng ◽

Weiwei Sun

Keyword(s):

Network Architecture ◽

Spatial Information ◽

Hyperspectral Image ◽

Hyperspectral Data ◽

Data Sets ◽

Learning Networks ◽

Hyperspectral Image Classification ◽

Residual Network ◽

Noise Interference ◽

Spectral Bands

Jointly using spectral and spatial information has become a mainstream strategy in the field of hyperspectral image (HSI) processing, especially for classification. However, due to the existence of noisy or correlated spectral bands in the spectral domain and inhomogeneous pixels in the spatial neighborhood, HSI classification results are often degraded and unsatisfactory. Motivated by the attention mechanism, this paper proposes a spatial–spectral squeeze-and-excitation (SSSE) module to adaptively learn the weights for different spectral bands and for different neighboring pixels. The SSSE structure can suppress or motivate features at a certain position, which can effectively resist noise interference and improve the classification results. Furthermore, we embed several SSSE modules into a residual network architecture and generate an SSSE-based residual network (SSSERN) model for HSI classification. The proposed SSSERN method is compared with several existing deep learning networks on two benchmark hyperspectral data sets. Experimental results demonstrate the effectiveness of our proposed network.

Download Full-text

Double-Branch Network with Pyramidal Convolution and Iterative Attention for Hyperspectral Image Classification

Remote Sensing ◽

10.3390/rs13071403 ◽

2021 ◽

Vol 13 (7) ◽

pp. 1403

Author(s):

Hao Shi ◽

Guo Cao ◽

Zixian Ge ◽

Youqiang Zhang ◽

Peng Fu

Keyword(s):

Hyperspectral Image ◽

State Of The Art ◽

Classification Performance ◽

Attention Mechanism ◽

Hyperspectral Images ◽

Hyperspectral Image Classification ◽

Excellent Performance ◽

Common Procedure ◽

Spatial Features ◽

First Choice

Deep-learning methods, especially convolutional neural networks (CNN), have become the first choice for hyperspectral image (HSI) classification to date. It is a common procedure that small cubes are cropped from hyperspectral images and then fed into CNNs. However, standard CNNs find it difficult to extract discriminative spectral–spatial features. How to obtain finer spectral–spatial features to improve the classification performance is now a hot topic of research. In this regard, the attention mechanism, which has achieved excellent performance in other computer vision, holds the exciting prospect. In this paper, we propose a double-branch network consisting of a novel convolution named pyramidal convolution (PyConv) and an iterative attention mechanism. Each branch concentrates on exploiting spectral or spatial features with different PyConvs, supplemented by the attention module for refining the feature map. Experimental results demonstrate that our model can yield competitive performance compared to other state-of-the-art models.

Download Full-text

ODPA-CNN: One Dimensional Parallel Atrous Convolution Neural Network for Band-Selective Hyperspectral Image Classification

Applied Sciences ◽

10.3390/app12010174 ◽

2021 ◽

Vol 12 (1) ◽

pp. 174

Author(s):

Byungjin Kang ◽

Inho Park ◽

Changmin Ok ◽

Sungho Kim

Keyword(s):

Neural Network ◽

Spatial Information ◽

Hyperspectral Image ◽

Convolution Neural Network ◽

Spectral Information ◽

Hyperspectral Image Classification ◽

One Dimensional ◽

Proposed Model ◽

3D Cnn ◽

2D And 3D

Recently, hyperspectral image (HSI) classification using deep learning has been actively studied using 2D and 3D convolution neural networks (CNN). However, they learn spatial information as well as spectral information. These methods can increase the accuracy of classification, but do not only focus on the spectral information, which is a big advantage of HSI. In addition, the 1D-CNN, which learns only pure spectral information, has limitations because it uses adjacent spectral information. In this paper, we propose a One Dimensional Parellel Atrous Convolution Neural Network (ODPA-CNN) that learns not only adjacent spectral information for HSI classification, but also spectral information from a certain distance. It extracts features in parallel to account for bands of varying distances. The proposed method excludes spatial information such as the shape of an object and performs HSI classification only with spectral information about the material of the object. Atrous convolution is not a convolution of adjacent spectral information, but a convolution between spectral information separated by a certain distance. We compare the proposed model with various datasets to the other models. We also test with the data we have taken ourselves. Experimental results show a higher performance than some 3D-CNN models and other 1D-CNN methods. In addition, using datasets to which random space is applied, the vulnerabilities of 3D-CNN are identified, and the proposed model is shown to be robust to datasets with little spatial information.

Download Full-text

Joint Alternate Small Convolution and Feature Reuse for Hyperspectral Image Classification

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi7090349 ◽

2018 ◽

Vol 7 (9) ◽

pp. 349 ◽

Cited By ~ 6

Author(s):

Hongmin Gao ◽

Yao Yang ◽

Chenming Li ◽

Hui Zhou ◽

Xiaoyu Qu

Keyword(s):

Precision Agriculture ◽

Strong Correlation ◽

Spatial Information ◽

Hyperspectral Image ◽

Spectral Feature ◽

Composite Layer ◽

Hyperspectral Data ◽

Spectral Information ◽

Model Parameters ◽

High Level

A hyperspectral image (HSI) contains fine and rich spectral information and spatial information of ground objects, which has great potential in applications. It is also widely used in precision agriculture, marine monitoring, military reconnaissance and many other fields. In recent years, a convolutional neural network (CNN) has been successfully used in HSI classification and has provided it with outstanding capacity for improving classification effects. To get rid of the bondage of strong correlation among bands for HSI classification, an effective CNN architecture is proposed for HSI classification in this work. The proposed CNN architecture has several distinct advantages. First, each 1D spectral vector that corresponds to a pixel in an HSI is transformed into a 2D spectral feature matrix, thereby emphasizing the difference among samples. In addition, this architecture can not only weaken the influence of strong correlation among bands on classification, but can also fully utilize the spectral information of hyperspectral data. Furthermore, a 1 × 1 convolutional layer is adopted to better deal with HSI information. All the convolutional layers in the proposed CNN architecture are composed of small convolutional kernels. Moreover, cascaded composite layers of the architecture consist of 1 × 1 and 3 × 3 convolutional layers. The inputs and outputs of each composite layer are stitched as the inputs of the next composite layer, thereby accomplishing feature reuse. This special module with joint alternate small convolution and feature reuse can extract high-level features from hyperspectral data meticulously and comprehensively solve the overfitting problem to an extent, in order to obtain a considerable classification effect. Finally, global average pooling is used to replace the traditional fully connected layer to reduce the model parameters and extract high-dimensional features from the hyperspectral data at the end of the architecture. Experimental results on three benchmark HSI datasets show the high classification accuracy and effectiveness of the proposed method.

Download Full-text