Automatic Design of Convolutional Neural Network for Hyperspectral Image Classification

Chen, Yushi; Zhu, Kaiqiang; Zhu, Lin; He, Xin; Ghamisi, Pedram; Benediktsson, Jón Atli

doi:10.1109/tgrs.2019.2910603

Cited by 168 publications

(81 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This model takes advantage of both determination-point-process (DPP) based diversity-promoting deep metrics and multi-scale features for effective HSI classification. Chen et al [45] aimed at the problem that the handcraft network structure can not adapt well to different data sets, they proposed the automatic CNN models called 1-D auto-CNN and 3-D auto-CNN for HSI classification. Firstly, a search algorithm based on gradient descent is used to efficiently find the best network structure to evaluate the performance on the validation set.…”

Section: Hyperspectral Image Classification Methods Based On 2d-3d Cnnmentioning

confidence: 99%

Hyperspectral Image Classification Method Based on 2D–3D CNN and Multibranch Feature Fusion

Cao

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

112

View full text Add to dashboard Cite

The emergence of convolutional neural network (CNN) has greatly promoted the development of hyperspectral image (HSI) classification technology. However, the acquisition of HSI is difficult. Lack of training samples is the primary cause of low classification performance. Traditional CNN-based methods mainly use 2D CNN for feature extraction, which make interband correlations of HSIs underutilized. 3D CNN extracts the joint-spectral-spatial information representation, but it depends on a more complex model. Also, too deep or too shallow network cannot extract the image features well. To tackle these issues, we propose an HSI classification method based on 2D-3D CNN and multi-branch feature fusion. We first combine 2D CNN and 3D CNN to extract image features. Then, by means of the multi-branch neural network, three kinds of features from shallow to deep are extracted and fused in the spectral dimension. Finally, the fused features are passed into several fully connected layers and a softmax layer to obtain the classification results. In addition, our network model utilizes the state-of-the-art activation function Mish to further improve the classification performance. Our experimental results, conducted on four widely used HSI data sets, indicate that the proposed method achieves better performance than the existing alternatives.

show abstract

Section: Hyperspectral Image Classification Methods Based On 2d-3d Cnnmentioning

confidence: 99%

Hyperspectral Image Classification Method Based on 2D–3D CNN and Multibranch Feature Fusion

Cao

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

112

View full text Add to dashboard Cite

show abstract

“…Due to the advancement of information technology, more data is within the reach of researchers. The data-driven approaches have found their way into various fields including signal processing [18], control systems [19][20][21][22] and especially vision tasks [23][24][25][26][27]. In particular, the deep learning-based method has stood out among the data-driven approaches.…”

Section: Data-driven Approachesmentioning

confidence: 99%

An Image Compression Method for Video Surveillance System in Underground Mines Based on Residual Networks and Discrete Wavelet Transform

Zhang

Chen

et al. 2019

Electronics

View full text Add to dashboard Cite

Video surveillance systems play an important role in underground mines. Providing clear surveillance images is the fundamental basis for safe mining and disaster alarming. It is of significance to investigate image compression methods since the underground wireless channels only allow low transmission bandwidth. In this paper, we propose a new image compression method based on residual networks and discrete wavelet transform (DWT) to solve the image compression problem. The residual networks are used to compose the codec network. Further, we propose a novel loss function named discrete wavelet similarity (DW-SSIM) loss to train the network. Because the information of edges in the image is exposed through DWT coefficients, the proposed network can learn to preserve the edges better. Experiments show that the proposed method has an edge over the methods being compared in regards to the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), particularly at low compression ratios. Tests on noise-contaminated images also demonstrate the noise robustness of the proposed method. Our main contribution is that the proposed method is able to compress images at relatively low compression ratios while still preserving sharp edges, which suits the harsh wireless communication environment in underground mines.

show abstract

“…Hypersepctral imaging technique captures the image data with a large number of consecutive narrow bands spanning the visible-to-infrared spectrums (Chen et al 2019;Zhang, Li, and Du 2019). Owing to differences in reflectivity for different materials under different electromagnetic spectrums, this technique is very effective to discriminate the composition of material and has been widely used in the fields of agriculture, forestry, geology, oceanography, meteorology, hydrology, military, environmental monitoring (Cao et al 2019;Liu et al 2017;Pan et al 2019;Yakovliev et al 2019;Chen, Xiao, and Li 2016).…”

Section: Introductionmentioning

confidence: 99%

Review on graph learning for dimensionality reduction of hyperspectral image

Zhang

Luo

2020

Geo-spatial Information Science

View full text Add to dashboard Cite

Graph learning is an effective manner to analyze the intrinsic properties of data. It has been widely used in the fields of dimensionality reduction and classification for data. In this paper, we focus on the graph learning-based dimensionality reduction for a hyperspectral image. Firstly, we review the development of graph learning and its application in a hyperspectral image. Then, we mainly discuss several representative graph methods including two manifold learning methods, two sparse graph learning methods, and two hypergraph learning methods. For manifold learning, we analyze neighborhood preserving embedding and locality preserving projections which are two classic manifold learning methods and can be transformed into the form of a graph. For sparse graph, we introduce sparsity preserving graph embedding and sparse graph-based discriminant analysis which can adaptively reveal data structure to construct a graph. For hypergraph learning, we review binary hypergraph and discriminant hyper-Laplacian projection which can represent the high-order relationship of data.

show abstract

Automatic Design of Convolutional Neural Network for Hyperspectral Image Classification

Cited by 168 publications

References 54 publications

Hyperspectral Image Classification Method Based on 2D–3D CNN and Multibranch Feature Fusion

Hyperspectral Image Classification Method Based on 2D–3D CNN and Multibranch Feature Fusion

An Image Compression Method for Video Surveillance System in Underground Mines Based on Residual Networks and Discrete Wavelet Transform

Review on graph learning for dimensionality reduction of hyperspectral image

Contact Info

Product

Resources

About