A Robust Sparse Representation Model for Hyperspectral Image Classification

Zhang, Hongyan; Pižurica, Aleksandra

doi:10.3390/s17092087

Cited by 24 publications

(11 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HSIs not only record the spectrum of materials in the spectral domain but also capture the distribution of ground-objects in the spatial domain. As the distribution of ground materials typically shows some continuity, HSIs are composed of various nearly homogeneous regions made of pixels that belong to the same class with a very high probability [34][35][36][37][38][39][40]. For this reason, spectral signatures of pixels within small local regions are typically very similar.…”

Section: The Ssc-tv Modelmentioning

confidence: 99%

Sketch-Based Subspace Clustering of Hyperspectral Images

Zhang

Pižurica

2020

Remote Sensing

Self Cite

View full text Add to dashboard Cite

Sparse subspace clustering (SSC) techniques provide the state-of-the-art in clustering of hyperspectral images (HSIs). However, their computational complexity hinders their applicability to large-scale HSIs. In this paper, we propose a large-scale SSC-based method, which can effectively process large HSIs while also achieving improved clustering accuracy compared to the current SSC methods. We build our approach based on an emerging concept of sketched subspace clustering, which was to our knowledge not explored at all in hyperspectral imaging yet. Moreover, there are only scarce results on any large-scale SSC approaches for HSI. We show that a direct application of sketched SSC does not provide a satisfactory performance on HSIs but it does provide an excellent basis for an effective and elegant method that we build by extending this approach with a spatial prior and deriving the corresponding solver. In particular, a random matrix constructed by the Johnson-Lindenstrauss transform is first used to sketch the self-representation dictionary as a compact dictionary, which significantly reduces the number of sparse coefficients to be solved, thereby reducing the overall complexity. In order to alleviate the effect of noise and within-class spectral variations of HSIs, we employ a total variation constraint on the coefficient matrix, which accounts for the spatial dependencies among the neighbouring pixels. We derive an efficient solver for the resulting optimization problem, and we theoretically prove its convergence property under mild conditions. The experimental results on real HSIs show a notable improvement in comparison with the traditional SSC-based methods and the state-of-the-art methods for clustering of large-scale images.

show abstract

Section: The Ssc-tv Modelmentioning

confidence: 99%

Sketch-Based Subspace Clustering of Hyperspectral Images

Zhang

Pižurica

2020

Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…All the experimental results are assessed by the overall accuracy (OA), average accuracy (AA), and kappa statistics (k) [ 35 ]. In order to avoid the effects induced by the selection of training samples, ten independent Monte Karlo runs are performed and OA, AA, and k are all averaged by ten runs.…”

Section: Resultsmentioning

confidence: 99%

“…It tends to produce solutions where the adjacent pixels are likely to belong to the same class [ 3 ]. The MLL prior has been widely used in image segmentation problems [ 31 , 32 , 33 , 34 ] and is a generalization of the Ising model [ 35 , 36 , 37 ]. It can be formulated as:

where

is a tunable parameter controlling the degree of smoothness, Z is a normalization constant for the density, and

is the unit impulse function.…”

Section: Methodsmentioning

confidence: 99%

Linear vs. Nonlinear Extreme Learning Machine for Spectral-Spatial Classification of Hyperspectral Images

Cao

Yang

Ren

et al. 2017

Sensors

View full text Add to dashboard Cite

As a new machine learning approach, the extreme learning machine (ELM) has received much attention due to its good performance. However, when directly applied to hyperspectral image (HSI) classification, the recognition rate is low. This is because ELM does not use spatial information, which is very important for HSI classification. In view of this, this paper proposes a new framework for the spectral-spatial classification of HSI by combining ELM with loopy belief propagation (LBP). The original ELM is linear, and the nonlinear ELMs (or Kernel ELMs) are an improvement of linear ELM (LELM). However, based on lots of experiments and much analysis, it is found that the LELM is a better choice than nonlinear ELM for the spectral-spatial classification of HSI. Furthermore, we exploit the marginal probability distribution that uses the whole information in the HSI and learns such a distribution using the LBP. The proposed method not only maintains the fast speed of ELM, but also greatly improves the accuracy of classification. The experimental results in the well-known HSI data sets, Indian Pines, and Pavia University, demonstrate the good performance of the proposed method.

show abstract

“…According to the authors, this provides robustness against the noise and low data quality attributed to the atmospheric distortions, but no experimental evidence was reported in the paper. The concept of classifying the super-pixels was also explored by Huang et al, who exploited a sparse representation model to achieve robustness against the mixed noise [78]. Duan et al proposed to employ relative total variation (commonly used for noise reduction) to extract multi-scale structural features [79].…”

Section: Dealing With Low-quality Datamentioning

confidence: 99%

Towards On-Board Hyperspectral Satellite Image Segmentation: Understanding Robustness of Deep Learning through Simulating Acquisition Conditions

Nalepa

Myller²,

Cwiek³

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Although hyperspectral images capture very detailed information about the scanned objects, their efficient analysis, transfer, and storage are still important practical challenges due to their large volume. Classifying and segmenting such imagery are the pivotal steps in virtually all applications, hence developing new techniques for these tasks is a vital research area. Here, deep learning has established the current state of the art. However, deploying large-capacity deep models on-board an Earth observation satellite poses additional technological challenges concerned with their memory footprints, energy consumption requirements, and robustness against varying-quality image data, with the last problem being under-researched. In this paper, we tackle this issue, and propose a set of simulation scenarios that reflect a range of atmospheric conditions and noise contamination that may ultimately happen on-board an imaging satellite. We verify their impact on the generalization capabilities of spectral and spectral-spatial convolutional neural networks for hyperspectral image segmentation. Our experimental analysis, coupled with various visualizations, sheds more light on the robustness of the deep models and indicate that specific noise distributions can significantly deteriorate their performance. Additionally, we show that simulating atmospheric conditions is key to obtaining the learners that generalize well over image data acquired in different imaging settings.

show abstract

A Robust Sparse Representation Model for Hyperspectral Image Classification

Cited by 24 publications

References 47 publications

Sketch-Based Subspace Clustering of Hyperspectral Images

Sketch-Based Subspace Clustering of Hyperspectral Images

Linear vs. Nonlinear Extreme Learning Machine for Spectral-Spatial Classification of Hyperspectral Images

Towards On-Board Hyperspectral Satellite Image Segmentation: Understanding Robustness of Deep Learning through Simulating Acquisition Conditions

Contact Info

Product

Resources

About