scholarly journals Hyperspectral Image Classification with Localized Graph Convolutional Filtering

2021 ◽  
Vol 13 (3) ◽  
pp. 526
Author(s):  
Shengliang Pu ◽  
Yuanfeng Wu ◽  
Xu Sun ◽  
Xiaotong Sun
Keyword(s):  
Hyperspectral Image ◽  
Principal Component ◽  
Representation Learning ◽  
Classification Performance ◽  
Hyperspectral Data ◽  
Spectral Graph Theory ◽  
Feature Reduction ◽  
Graph Representation ◽  
Novel Method ◽  
Local Graph

The nascent graph representation learning has shown superiority for resolving graph data. Compared to conventional convolutional neural networks, graph-based deep learning has the advantages of illustrating class boundaries and modeling feature relationships. Faced with hyperspectral image (HSI) classification, the priority problem might be how to convert hyperspectral data into irregular domains from regular grids. In this regard, we present a novel method that performs the localized graph convolutional filtering on HSIs based on spectral graph theory. First, we conducted principal component analysis (PCA) preprocessing to create localized hyperspectral data cubes with unsupervised feature reduction. These feature cubes combined with localized adjacent matrices were fed into the popular graph convolution network in a standard supervised learning paradigm. Finally, we succeeded in analyzing diversified land covers by considering local graph structure with graph convolutional filtering. Experiments on real hyperspectral datasets demonstrated that the presented method offers promising classification performance compared with other popular competitors.

scholarly journals A Novel Method to Predict Drug-Target Interactions Based on Large-Scale Graph Representation Learning

Cancers ◽  
2021 ◽  
Vol 13 (9) ◽  
pp. 2111
Author(s):  
Bo-Wei Zhao ◽  
Zhu-Hong You ◽  
Lun Hu ◽  
Zhen-Hao Guo ◽  
Lei Wang ◽  
...  
Keyword(s):  
Drug Target ◽  
Large Scale ◽  
Computational Models ◽  
Structural Information ◽  
Characteristic Curve ◽  
Representation Learning ◽  
Graph Representation ◽  
Convolutional Network ◽  
Novel Method

Identification of drug-target interactions (DTIs) is a significant step in the drug discovery or repositioning process. Compared with the time-consuming and labor-intensive in vivo experimental methods, the computational models can provide high-quality DTI candidates in an instant. In this study, we propose a novel method called LGDTI to predict DTIs based on large-scale graph representation learning. LGDTI can capture the local and global structural information of the graph. Specifically, the first-order neighbor information of nodes can be aggregated by the graph convolutional network (GCN); on the other hand, the high-order neighbor information of nodes can be learned by the graph embedding method called DeepWalk. Finally, the two kinds of feature are fed into the random forest classifier to train and predict potential DTIs. The results show that our method obtained area under the receiver operating characteristic curve (AUROC) of 0.9455 and area under the precision-recall curve (AUPR) of 0.9491 under 5-fold cross-validation. Moreover, we compare the presented method with some existing state-of-the-art methods. These results imply that LGDTI can efficiently and robustly capture undiscovered DTIs. Moreover, the proposed model is expected to bring new inspiration and provide novel perspectives to relevant researchers.

scholarly journals Deep Spectral Spatial Inverted Residual Network for Hyperspectral Image Classification

2021 ◽  
Vol 13 (21) ◽  
pp. 4472
Author(s):  
Tianyu Zhang ◽  
Cuiping Shi ◽  
Diling Liao ◽  
Liguo Wang
Keyword(s):  
Image Classification ◽  
Data Augmentation ◽  
Hyperspectral Image ◽  
Classification Performance ◽  
Hyperspectral Data ◽  
Hyperspectral Images ◽  
Hyperspectral Image Classification ◽  
Feature Maps ◽  
Spatial Feature ◽  
Global Context

Convolutional neural networks (CNNs) have been widely used in hyperspectral image classification in recent years. The training of CNNs relies on a large amount of labeled sample data. However, the number of labeled samples of hyperspectral data is relatively small. Moreover, for hyperspectral images, fully extracting spectral and spatial feature information is the key to achieve high classification performance. To solve the above issues, a deep spectral spatial inverted residuals network (DSSIRNet) is proposed. In this network, a data block random erasing strategy is introduced to alleviate the problem of limited labeled samples by data augmentation of small spatial blocks. In addition, a deep inverted residuals (DIR) module for spectral spatial feature extraction is proposed, which locks the effective features of each layer while avoiding network degradation. Furthermore, a global 3D attention module is proposed, which can realize the fine extraction of spectral and spatial global context information under the condition of the same number of input and output feature maps. Experiments are carried out on four commonly used hyperspectral datasets. A large number of experimental results show that compared with some state-of-the-art classification methods, the proposed method can provide higher classification accuracy for hyperspectral images.

scholarly journals Quality Metrics Evaluation of Hyperspectral Images

2014 ◽  
Vol XL-8 ◽  
pp. 1221-1226 ◽  
Author(s):  
A. K. Singh ◽  
H. V. Kumar ◽  
G. R. Kadambi ◽  
J. K. Kishore ◽  
J. Shuttleworth ◽  
...  
Keyword(s):  
Hyperspectral Image ◽  
Signal To Noise Ratio ◽  
Principal Component ◽  
Quality Parameters ◽  
Quality Metrics ◽  
Hyperspectral Data ◽  
Hyperspectral Images ◽  
Error Ratio ◽  
Classified Image

In this paper, the quality metrics evaluation on hyperspectral images has been presented using k-means clustering and segmentation. After classification the assessment of similarity between original image and classified image is achieved by measurements of image quality parameters. Experiments were carried out on four different types of hyperspectral images. Aerial and spaceborne hyperspectral images with different spectral and geometric resolutions were considered for quality metrics evaluation. Principal Component Analysis (PCA) has been applied to reduce the dimensionality of hyperspectral data. PCA was ultimately used for reducing the number of effective variables resulting in reduced complexity in processing. In case of ordinary images a human viewer plays an important role in quality evaluation. Hyperspectral data are generally processed by automatic algorithms and hence cannot be viewed directly by human viewers. Therefore evaluating quality of classified image becomes even more significant. An elaborate comparison is made between k-means clustering and segmentation for all the images by taking Peak Signal-to-Noise Ratio (PSNR), Mean Square Error (MSE), Maximum Squared Error, ratio of squared norms called L2RAT and Entropy. First four parameters are calculated by comparing the quality of original hyperspectral image and classified image. Entropy is a measure of uncertainty or randomness which is calculated for classified image. Proposed methodology can be used for assessing the performance of any hyperspectral image classification techniques.

Design and Development of EMG Conditioning System and Hand Gesture Recognition Based on Principal Component Analysis Feature Reduction Technique

2014 ◽  
pp. 304-320
Author(s):  
P. Geethanjali
Keyword(s):  
Principal Component Analysis ◽  
Principal Component ◽  
Digital Signal ◽  
Component Analysis ◽  
Classification Performance ◽  
Feature Reduction ◽  
Design And Development ◽  
Emg Signal ◽  
Knn Classifier ◽  
Power Line Interference

This chapter discusses design and development of a surface Electromyogram (EMG) signal detection and conditioning system along with the issues of gratuitous spurious signals such as power line interference, artifacts, etc., which make signals plausible. In order to construe the recognition of hand gestures from EMG signals, Time Domain (TD) and well as Autoregressive (AR) coefficients features are extracted. The extracted features are diminished using the Principal Component Analysis (PCA) to alleviate the burden of the classifier. A four-channel continuous EMG signal conditioning system is developed and EMG signals are acquired from 10 able-bodied subjects to classify the 6 unique movements of hand and wrist. The reduced statistical TD and AR features are used to classify the signal patterns through k Nearest Neighbour (kNN) as well as Neural Network (NN) classifier. Further, EMG signals acquired from a transradial amputee using 8-channel systems for the 6 amenable motions are also classified. Statistical Analysis of Variance (ANOVA) results on classification performance of able-bodied subject divulge that the performance TD-PCA features are more significant than the AR-PCA features. Further, no significant difference in the performance of NN classifier and kNN classifier is construed with TD reduced features. Since the average classification error of kNN classifier with TD features is found to be less, kNN classifier is implemented in off-line using the TMS2407eZdsp digital signal controller to study the actuation of three low-power DC drives in the identification of intended motion with an able-bodied subject.

scholarly journals Kernel Joint Sparse Representation Based on Self-Paced Learning for Hyperspectral Image Classification

2019 ◽  
Vol 11 (9) ◽  
pp. 1114
Author(s):  
Sixiu Hu ◽  
Jiangtao Peng ◽  
Yingxiong Fu ◽  
Luoqing Li
Keyword(s):  
Sparse Representation ◽  
Hyperspectral Image ◽  
Feature Space ◽  
Classification Performance ◽  
Hyperspectral Data ◽  
Data Sets ◽  
Neighborhood Structure ◽  
Joint Sparse Representation ◽  
Negative Effect ◽  
Sparse Coefficient

By means of joint sparse representation (JSR) and kernel representation, kernel joint sparse representation (KJSR) models can effectively model the intrinsic nonlinear relations of hyperspectral data and better exploit spatial neighborhood structure to improve the classification performance of hyperspectral images. However, due to the presence of noisy or inhomogeneous pixels around the central testing pixel in the spatial domain, the performance of KJSR is greatly affected. Motivated by the idea of self-paced learning (SPL), this paper proposes a self-paced KJSR (SPKJSR) model to adaptively learn weights and sparse coefficient vectors for different neighboring pixels in the kernel-based feature space. SPL strateges can learn a weight to indicate the difficulty of feature pixels within a spatial neighborhood. By assigning small weights for unimportant or complex pixels, the negative effect of inhomogeneous or noisy neighboring pixels can be suppressed. Hence, SPKJSR is usually much more robust. Experimental results on Indian Pines and Salinas hyperspectral data sets demonstrate that SPKJSR is much more effective than traditional JSR and KJSR models.

scholarly journals An Effective Classification Scheme for Hyperspectral Image Based on Superpixel and Discontinuity Preserving Relaxation

2019 ◽  
Vol 11 (10) ◽  
pp. 1149 ◽  
Author(s):  
Fuding Xie ◽  
Cunkuan Lei ◽  
Jun Yang ◽  
Cui Jin
Keyword(s):  
Spatial Structure ◽  
Classification Scheme ◽  
Hyperspectral Image ◽  
Principal Component ◽  
Hyperspectral Data ◽  
Novel Technique ◽  
Classification Framework ◽  
Comparison Results ◽  
Simple Linear Iterative Clustering ◽  
Land Covers

Hyperspectral image (HSI) classification is one of the most active topics in remote sensing. However, it is still a nontrivial task to classify the hyperspectral data accurately, since HSI always suffers from a large number of noise pixels, the complexity of the spatial structure of objects and the spectral similarity between different objects. In this study, an effective classification scheme for hyperspectral image based on superpixel and discontinuity preserving relaxation (DPR) is proposed to discriminate land covers of interest. A novel technique for measuring the similarity of a pair of pixels in HSI is suggested to improve the simple linear iterative clustering (SLIC) algorithm. Unlike the existing application of SLIC technique to HSI, the improved SLIC algorithm can be directly used to segment HSI into superpixels without using principal component analysis in advance, and is free of parameters. Furthermore, the proposed three-step classification scheme explores how to effectively use the global spectral information and local spatial structure of hyperspectral data for HSI classification. Compared with the existing two-step classification framework, the use of DPR technology in preprocessing significantly improves the classification accuracy. The effectiveness of the proposed method is verified on three public real hyperspectral datasets. The comparison results of several competitive methods show the superiority of this scheme.

scholarly journals Prediction of Sweet Corn Seed Germination Based on Hyperspectral Image Technology and Multivariate Data Regression

Sensors ◽  
2020 ◽  
Vol 20 (17) ◽  
pp. 4744
Author(s):  
Huawei Cui ◽  
Zhishang Cheng ◽  
Peng Li ◽  
Aimin Miao
Keyword(s):  
Seed Germination ◽  
Correlation Coefficient ◽  
Root Length ◽  
Sweet Corn ◽  
Hyperspectral Image ◽  
Principal Component Regression ◽  
Principal Component ◽  
Hyperspectral Data ◽  
Seed Vigor ◽  
Seedling Length

Vigor identification in sweet corn seeds is important for seed germination, crop yield, and quality. In this study, hyperspectral image (HSI) technology integrated with germination tests was applied for feature association analysis and germination performance prediction of sweet corn seeds. In this study, 89 sweet corn seeds (73 for training and the other 16 for testing) were studied and hyperspectral imaging at the spectral range of 400–1000 nm was applied as a nondestructive and accurate technique to identify seed vigor. The root length and seedling length which represent the seed vigor were measured, and principal component regression (PCR), partial least squares (PLS), and kernel principal component regression (KPCR) were used to establish the regression relationship between the hyperspectral feature of seeds and the germination results. Specifically, the relevant characteristic band associated with seed vigor based on the highest correlation coefficient (HCC) was constructed for optimal wavelength selection. The hyperspectral data features were selected by genetic algorithm (GA), successive projections algorithm (SPA), and HCC. The results indicated that the hyperspectral data features obtained based on the HCC method have better prediction results on the seedling length and root length than SPA and GA. By comparing the regression results of KPCR, PCR, and PLS, it can be concluded that the hyperspectral method can predict the root length with a correlation coefficient of 0.7805. The prediction results of different feature selection and regression algorithms for the seedling length were up to 0.6074. The results indicated that, based on hyperspectral technology, the prediction of seedling root length was better than that of seed length.

scholarly journals LOCAL BINARY GRAPH FEATURE REDUCTION FOR THREE-DIMENSIONAL GABOR FILTER BASED HYPERSPECTRAL IMAGE CLASSIFICATION

2019 ◽  
Vol XLII-4/W18 ◽  
pp. 285-291 ◽  
Author(s):  
M. Darvishnezhad ◽  
H. Ghassemian ◽  
M. Imani
Keyword(s):  
Image Classification ◽  
Hyperspectral Image ◽  
Gabor Filter ◽  
Three Dimensional ◽  
Gabor Filters ◽  
Feature Reduction ◽  
Hyperspectral Image Classification ◽  
Data Set ◽  
Spatial Features ◽  
Novel Method

Abstract. One of the challenges of the hyperspectral image classification is the fusing spectral and spatial features. There are several methods for fusing features in hyperspectral image classification. Three-Dimensional Gabor Filters are the best method to extract spectral and spatial features simultaneously. However, one of the problems with using the 3D Gabor filter is the high number of extracted features. In this paper, to reducing extracted features from 3D-Gabor filters and increasing the classification accuracy in hyperspectral images, a novel method named Local Binary Graph (LBG) is used. The LBG method uses a local graph to solve the optimization problem, which maps each pixel to the reduced dimension image and improves the McNemar test result in comparison with the existing methods. Finally, the result of the proposed method achieved 96.2% and 92.6% overall accuracy for Pavia University and Indian Pines data set, respectively.

scholarly journals Segment-Based Clustering of Hyperspectral Images Using Tree-Based Data Partitioning Structures

Algorithms ◽  
2020 ◽  
Vol 13 (12) ◽  
pp. 330
Author(s):  
Mohamed Ismail ◽  
Milica Orlandić
Keyword(s):  
Large Scale ◽  
Clustering Algorithm ◽  
Hyperspectral Image ◽  
Principal Component ◽  
Data Partitioning ◽  
Hyperspectral Data ◽  
Compact Representation ◽  
Parametric Modelling ◽  
Ground Truth Data ◽  
Binary Partition

Hyperspectral image classification has been increasingly used in the field of remote sensing. In this study, a new clustering framework for large-scale hyperspectral image (HSI) classification is proposed. The proposed four-step classification scheme explores how to effectively use the global spectral information and local spatial structure of hyperspectral data for HSI classification. Initially, multidimensional Watershed is used for pre-segmentation. Region-based hierarchical hyperspectral image segmentation is based on the construction of Binary partition trees (BPT). Each segmented region is modeled while using first-order parametric modelling, which is then followed by a region merging stage using HSI regional spectral properties in order to obtain a BPT representation. The tree is then pruned to obtain a more compact representation. In addition, principal component analysis (PCA) is utilized for HSI feature extraction, so that the extracted features are further incorporated into the BPT. Finally, an efficient variant of k-means clustering algorithm, called filtering algorithm, is deployed on the created BPT structure, producing the final cluster map. The proposed method is tested over eight publicly available hyperspectral scenes with ground truth data and it is further compared with other clustering frameworks. The extensive experimental analysis demonstrates the efficacy of the proposed method.

scholarly journals Deep Convolutional Capsule Network for Hyperspectral Image Spectral and Spectral-Spatial Classification

2019 ◽  
Vol 11 (3) ◽  
pp. 223 ◽  
Author(s):  
Kaiqiang Zhu ◽  
Yushi Chen ◽  
Pedram Ghamisi ◽  
Xiuping Jia ◽  
Jón Atli Benediktsson
Keyword(s):  
Network Architecture ◽  
Hyperspectral Image ◽  
Principal Component ◽  
Hyperspectral Data ◽  
Support Vector ◽  
Data Sets ◽  
Spatial Classification ◽  
Training Samples ◽  
Vector Machines ◽  
Kernel Support Vector Machines

Capsule networks can be considered to be the next era of deep learning and have recently shown their advantages in supervised classification. Instead of using scalar values to represent features, the capsule networks use vectors to represent features, which enriches the feature presentation capability. This paper introduces a deep capsule network for hyperspectral image (HSI) classification to improve the performance of the conventional convolutional neural networks (CNNs). Furthermore, a modification of the capsule network named Conv-Capsule is proposed. Instead of using full connections, local connections and shared transform matrices, which are the core ideas of CNNs, are used in the Conv-Capsule network architecture. In Conv-Capsule, the number of trainable parameters is reduced compared to the original capsule, which potentially mitigates the overfitting issue when the number of available training samples is limited. Specifically, we propose two schemes: (1) A 1D deep capsule network is designed for spectral classification, as a combination of principal component analysis, CNN, and the Conv-Capsule network, and (2) a 3D deep capsule network is designed for spectral-spatial classification, as a combination of extended multi-attribute profiles, CNN, and the Conv-Capsule network. The proposed classifiers are tested on three widely-used hyperspectral data sets. The obtained results reveal that the proposed models provide competitive results compared to the state-of-the-art methods, including kernel support vector machines, CNNs, and recurrent neural network.

Sign in / Sign up

Export Citation Format

Share Document