scholarly journals Sparse Subspace Clustering in Hyperspectral Images using Incomplete Pixels

TecnoLógicas ◽  
2019 ◽  
Vol 22 (46) ◽  
pp. 1-14 ◽  
Author(s):  
Jorge Luis Bacca ◽  
Henry Arguello
Keyword(s):  
Spatial Information ◽  
Hyperspectral Image ◽  
Subspace Clustering ◽  
Unsupervised Classification ◽  
Hyperspectral Images ◽  
Image Clustering ◽  
Spectral Signature ◽  
Spectral Image ◽  
Sparse Subspace Clustering

Spectral image clustering is an unsupervised classification method which identifies distributions of pixels using spectral information without requiring a previous training stage. The sparse subspace clustering-based methods (SSC) assume that hyperspectral images lie in the union of multiple low-dimensional subspaces.  Using this, SSC groups spectral signatures in different subspaces, expressing each spectral signature as a sparse linear combination of all pixels, ensuring that the non-zero elements belong to the same class. Although these methods have shown good accuracy for unsupervised classification of hyperspectral images, the computational complexity becomes intractable as the number of pixels increases, i.e. when the spatial dimension of the image is large. For this reason, this paper proposes to reduce the number of pixels to be classified in the hyperspectral image, and later, the clustering results for the missing pixels are obtained by exploiting the spatial information. Specifically, this work proposes two methodologies to remove the pixels, the first one is based on spatial blue noise distribution which reduces the probability to remove cluster of neighboring pixels, and the second is a sub-sampling procedure that eliminates every two contiguous pixels, preserving the spatial structure of the scene. The performance of the proposed spectral image clustering framework is evaluated in three datasets showing that a similar accuracy is obtained when up to 50% of the pixels are removed, in addition, it is up to 7.9 times faster compared to the classification of the data sets without incomplete pixels.

scholarly journals Overview of Hyperspectral Image Classification

2020 ◽  
Vol 2020 ◽  
pp. 1-13 ◽  
Author(s):  
Wenjing Lv ◽  
Xiaofei Wang
Keyword(s):  
Supervised Classification ◽  
Hyperspectral Image ◽  
Unsupervised Classification ◽  
Hyperspectral Images ◽  
Hyperspectral Image Classification ◽  
Remote Sensing Technology ◽  
Sensing Technology ◽  
Good Classification ◽  
Semisupervised Classification

With the development of remote sensing technology, the application of hyperspectral images is becoming more and more widespread. The accurate classification of ground features through hyperspectral images is an important research content and has attracted widespread attention. Many methods have achieved good classification results in the classification of hyperspectral images. This paper reviews the classification methods of hyperspectral images from three aspects: supervised classification, semisupervised classification, and unsupervised classification.

scholarly journals Efficient Unsupervised Classification of Hyperspectral Images Using Voronoi Diagrams and Strong Patterns

Sensors ◽  
2020 ◽  
Vol 20 (19) ◽  
pp. 5684
Author(s):  
Laura Bianca Bilius ◽  
Ştefan Gheorghe Pentiuc
Keyword(s):  
Heuristic Method ◽  
Voronoi Diagrams ◽  
Ground Truth ◽  
Unsupervised Classification ◽  
Hyperspectral Data ◽  
Hyperspectral Images ◽  
Spectral Signature ◽  
Area Of Interest ◽  
Parafac Decomposition

Hyperspectral images (HSIs) are a powerful tool to classify the elements from an area of interest by their spectral signature. In this paper, we propose an efficient method to classify hyperspectral data using Voronoi diagrams and strong patterns in the absence of ground truth. HSI processing consumes a great deal of computing resources because HSIs are represented by large amounts of data. We propose a heuristic method that starts by applying Parafac decomposition for reduction and to construct the abundances matrix. Furthermore, the representative nodes from the abundances map are searched for. A multi-partition of these nodes is found, and based on this, strong patterns are obtained. Then, based on the hierarchical clustering of strong patterns, an optimum partition is found. After strong patterns are labeled, we construct the Voronoi diagram to extend the classification to the entire HSI.

scholarly journals UNLABELED SELECTED SAMPLES IN FEATURE EXTRACTION FOR CLASSIFICATION OF HYPERSPECTRAL IMAGES WITH LIMITED TRAINING SAMPLES

2015 ◽  
Vol XL-1-W5 ◽  
pp. 411-416
Author(s):  
A. Kianisarkaleh ◽  
H. Ghassemian ◽  
F. Razzazi
Keyword(s):  
Feature Extraction ◽  
Spatial Information ◽  
Hyperspectral Image ◽  
Sample Selection ◽  
Extraction Methods ◽  
Hyperspectral Images ◽  
Training Samples ◽  
Samples Selection ◽  
Limited Training Samples

Feature extraction plays a key role in hyperspectral images classification. Using unlabeled samples, often unlimitedly available, unsupervised and semisupervised feature extraction methods show better performance when limited number of training samples exists. This paper illustrates the importance of selecting appropriate unlabeled samples that used in feature extraction methods. Also proposes a new method for unlabeled samples selection using spectral and spatial information. The proposed method has four parts including: PCA, prior classification, posterior classification and sample selection. As hyperspectral image passes these parts, selected unlabeled samples can be used in arbitrary feature extraction methods. The effectiveness of the proposed unlabeled selected samples in unsupervised and semisupervised feature extraction is demonstrated using two real hyperspectral datasets. Results show that through selecting appropriate unlabeled samples, the proposed method can improve the performance of feature extraction methods and increase classification accuracy.

Spatial distribution preserving-based sparse subspace clustering for hyperspectral image

2019 ◽  
Vol 17 (02) ◽  
pp. 1940010
Author(s):  
Yiyang Ding ◽  
Anyong Qin ◽  
Zhaowei Shang ◽  
Jiye Qian
Keyword(s):  
Spatial Distribution ◽  
Spatial Information ◽  
Hyperspectral Image ◽  
Subspace Clustering ◽  
Original Data ◽  
Coefficient Matrix ◽  
Intrinsic Structure ◽  
Affine Subspaces ◽  
Novel Approach ◽  
Sparse Subspace Clustering

The high dimensionality and heterogeneity of the hyperspectral image (HSI) make a challenge to the application of machine learning methods, such as sparse subspace clustering (SSC). SSC is designed to represent data as an union of affine subspaces, while it cannot capture the latent structure of the given data. In Mosers theory, the distribution can represent the intrinsic structure efficiently. Hence, we propose a novel approach called spatial distribution preserving-based sparse subspace clustering (SSC-SDP) in this paper for HSI data, which can help sparse representation preserve the underlying manifold structure. Specifically, the density constraint is added by minimizing the inconsistency of the densities estimated in the HSI data and the corresponding sparse coefficient matrix. In addition, we incorporate spatial information into the density estimation of the original data, and the optimization solution based on alternating direction method of multipliers (ADMM) is devised. Three HSI data sets are conducted to evaluate the performance of our SSC-SDP compared with other state-of-art algorithms.

Hyperspectral Image Classification Using Spatial and Edge Features Based on Deep Learning

2019 ◽  
Vol 33 (09) ◽  
pp. 1954027 ◽  
Author(s):  
Dexiang Zhang ◽  
Jingzhong Kang ◽  
Lina Xun ◽  
Yu Huang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Spatial Information ◽  
Hyperspectral Image ◽  
Bilateral Filter ◽  
Convolution Neural Network ◽  
Hyperspectral Images ◽  
Data Sets ◽  
Edge Information

In recent years, deep learning has been widely used in the classification of hyperspectral images and good results have been achieved. But it is easy to ignore the edge information of the image when using the spatial features of hyperspectral images to carry out the classification experiments. In order to make full use of the advantages of convolution neural network (CNN), we extract the spatial information with the method of minimum noise fraction (MNF) and the edge information by bilateral filter. The combination of the two kinds of information not only increases the useful information but also effectively removes part of the noise. The convolution neural network is used to extract features and classify for hyperspectral images on the basis of this fused information. In addition, this paper also uses another kind of edge-filtering method to amend the final classification results for a better accuracy. The proposed method was tested on three public available data sets: the University of Pavia, the Salinas, and the Indian Pines. The competitive results indicate that our approach can realize a classification of different ground targets with a very high accuracy.

Hyperspectral Image Classification with Kernels

2011 ◽  
pp. 374-398 ◽  
Author(s):  
Lorenzo Bruzzone ◽  
Luis Gomez-Chova ◽  
Mattia Marconcini ◽  
Gustavo Camps-Valls
Keyword(s):  
Kernel Methods ◽  
Hyperspectral Image ◽  
Hyperspectral Images ◽  
Spectral Signature ◽  
Quality Of Data ◽  
Critical Problems ◽  
Improved Accuracy ◽  
Spectral Channels

The information contained in hyperspectral images allows the characterization, identification, and classification of land covers with improved accuracy and robustness. However, several critical problems should be considered in the classification of hyperspectral images, among which are (a) the high number of spectral channels, (b) the spatial variability of the spectral signature, (c) the high cost of true sample labeling, and (d) the quality of data. Recently, kernel methods have offered excellent results in this context. This chapter reviews the state-of-the-art hyperspectral image classifiers, presents two recently proposed kernel-based approaches, and systematically discusses the specific needs and demands of this field.

Spectral-Spatial Sparse Subspace Clustering Based on Three-Dimensional Edge-Preserving Filtering for Hyperspectral Image

2019 ◽  
Vol 33 (03) ◽  
pp. 1955003 ◽  
Author(s):  
Ailin Li ◽  
Anyong Qin ◽  
Zhaowei Shang ◽  
Yuan Yan Tang
Keyword(s):  
Spatial Information ◽  
Hyperspectral Image ◽  
Three Dimensional ◽  
Subspace Clustering ◽  
Coefficient Matrix ◽  
Simple Method ◽  
Edge Preserving ◽  
Sparse Coefficient ◽  
Similarity Graph ◽  
Sparse Subspace Clustering

Integrating spatial information into the sparse subspace clustering (SSC) models for hyperspectral images (HSIs) is an effective way to improve clustering accuracy. Since HSI is a three-dimensional (3D) cube datum, 3D spectral-spatial filtering becomes a simple method for extracting the spectral-spatial information. In this paper, a novel spectral-spatial SSC framework based on 3D edge-preserving filtering (EPF) is proposed to improve the clustering accuracy of HSI. First, the initial sparse coefficient matrix is obtained in the sparse representation process of the classical SSC model. Then, a 3D EPF is conducted on the initial sparse coefficient matrix to obtain a more accurate coefficient matrix by solving an optimization problem based on ADMM, which is used to build the similarity graph. Finally, the clustering result of HSI data is achieved by applying the spectral clustering algorithm to the similarity graph. Specifically, the filtered matrix can not only capture the spectral-spatial information but the intensity differences. The experimental results on three real-world HSI datasets demonstrated that the potential of including the proposed 3D EPF into the SSC framework can improve the clustering accuracy.

scholarly journals Cluster-Wise Weighted NMF for Hyperspectral Images Unmixing with Imbalanced Data

2021 ◽  
Vol 13 (2) ◽  
pp. 268
Author(s):  
Xiaochen Lv ◽  
Wenhong Wang ◽  
Hongfu Liu
Keyword(s):  
Spatial Information ◽  
Hyperspectral Image ◽  
Imbalanced Data ◽  
Reconstruction Error ◽  
Hyperspectral Data ◽  
Weight Matrix ◽  
Hyperspectral Images ◽  
Mixed Data ◽  
Sparsity Constraints ◽  
Additional Constraints

Hyperspectral unmixing is an important technique for analyzing remote sensing images which aims to obtain a collection of endmembers and their corresponding abundances. In recent years, non-negative matrix factorization (NMF) has received extensive attention due to its good adaptability for mixed data with different degrees. The majority of existing NMF-based unmixing methods are developed by incorporating additional constraints into the standard NMF based on the spectral and spatial information of hyperspectral images. However, they neglect to exploit the nature of imbalanced pixels included in the data, which may cause the pixels mixed with imbalanced endmembers to be ignored, and thus the imbalanced endmembers generally cannot be accurately estimated due to the statistical property of NMF. To exploit the information of imbalanced samples in hyperspectral data during the unmixing procedure, in this paper, a cluster-wise weighted NMF (CW-NMF) method for the unmixing of hyperspectral images with imbalanced data is proposed. Specifically, based on the result of clustering conducted on the hyperspectral image, we construct a weight matrix and introduce it into the model of standard NMF. The proposed weight matrix can provide an appropriate weight value to the reconstruction error between each original pixel and the reconstructed pixel in the unmixing procedure. In this way, the adverse effect of imbalanced samples on the statistical accuracy of NMF is expected to be reduced by assigning larger weight values to the pixels concerning imbalanced endmembers and giving smaller weight values to the pixels mixed by majority endmembers. Besides, we extend the proposed CW-NMF by introducing the sparsity constraints of abundance and graph-based regularization, respectively. The experimental results on both synthetic and real hyperspectral data have been reported, and the effectiveness of our proposed methods has been demonstrated by comparing them with several state-of-the-art methods.

Sign in / Sign up

Export Citation Format

Share Document