Optimized Ensemble EMD-Based Spectral Features for Hyperspectral Image Classification

Abstract

Extracting essential features from massive bands is an important yet challenging issue in hyperspectral image (HSI) classification. Plenty of feature extraction techniques can be found in the literature but most of these methods rely on linear/stationary assumptions. This paper proposes an alternative methodology inspired by the ensemble empirical mode decomposition (EEMD) to gain spectral features of the HSI. To this end, two major aspects are involved: 1) the optimization problems are formulated in each sifting process and solved by the alternating direction method of multipliers (ADMM) algorithm to enhance the benefits of EEMD; 2) the intrinsic mode functions (IMFs) extracted by the optimized EEMD (OEEMD) are summed with appropriate weights automatically gained from the local Fisher discriminant analysis (LFDA). As a consequence, the constructed features (i.e., sum of the IMFs) can then be significantly classified by the state-of-the-art classifiers, i.e., k-nearest neighbor (k-NN) or support vector machine (SVM). Experiments on two benchmark HSIs validate that the extracted new features achieve higher classification rates as well as greater robustness to the choice of training samples compared with several generally acknowledged methods.Index Terms-Alternating direction method of multipliers (ADMM), classification, ensemble empirical mode decomposition (EEMD), hyperspectral image (HSI), local Fisher discriminant analysis (LFDA).

Customized Orthogonal Locality Preserving Projections With Soft-Margin Maximization for Face Recognition

“…Therefore, the NN classifier commonly used in the appearance-based face recognition approach may be insufficient, especially because it often fails with noisy high-dimensional samples and outliers. The SVM approach does not need to assume any hypothesis about the data distribution [32], [33], and has proven to be efficient in many classification applications, such as face recognition [24], [25], age estimation [12], [13], hyperspectral image classification [34], and other applications involving high-dimensional data. Therefore, to better handle these common classification problems in face recognition, we propose to perform classification of new face samples using the soft-margin soft-margin support vector machine (SVM2) method.…”

Effective Denoising and Classification of Hyperspectral Images Using Curvelet Transform and Singular Spectrum Analysis

“…Then, the CT-SSA method as well as its extended version CT-SSA-PCA, are further compared with some state-of-the-art spectral feature extraction techniques in subsection B, including PCA, LDA, NMF and ensemble EMD (EEMD) [55], to show the efficacy of the proposed methodology.…”

“…The NMF algorithm is realized by the NMF Matlab toolbox [58]. Additionally, as an upgraded version of EMD, EEMD is also included for comparison as it overcomes the drawback of EMD, and therefore it outperforms EMD [55]. Being the fundamental part of the Hilbert-Huang transform (HHT), EMD is a non-linear and non-stationary signal decomposition method that decomposes the signal into a finite number of intrinsic mode functions (IMFs) [59].…”

Segmented minimum noise fraction transformation for efficient feature extraction of hyperspectral images

“…In greater detail, the classification results are compared numerically [overall accuracy (OA) and average accuracy (AA)] and statistically [kappa coefficient (kappa)]. For the pixel classifiers, k is simply set to 1 in the k-NN, whereas the widely used OAO strategy and RBF kernel are chosen in the SVM, in which C is set as 60 for all experiments [10]. The k-NN classifier in the pattern recognition toolbox of MATLAB R2013a and the SVM classifier in the OSU-SVM toolbox, which is based on LibSVM [11], are utilized in this paper.…”