Limitations of Principal Components Analysis for Hyperspectral Target Recognition

By Saurabh Prasad, Lori Mann Bruce. IEEE Geoscience and Remote Sensing Letters, 2008

327 citations recorded by the source corpus.

Subfields: Analytical Chemistry, Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/49d98c479434c898dfe23a2d2d0450518bc0c9e8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Analytical Chemistry

1. Shutao Li
2. M. Dalla Mura
3. Xudong Kang
4. Leyuan Fang
5. Zebin Wu
6. Guangrui Zhao
7. Yuhui Zheng
8. Devis Tuia

In Computer Vision and Pattern Recognition

1. Leyuan Fang
2. Shutao Li
3. Devis Tuia
4. Xudong Kang
5. Alain Rakotomamonjy
6. M. Dalla Mura
7. Rémi Flamary
8. Jon Atli Benediktsson

In Media Technology

1. Shutao Li
2. Leyuan Fang
3. Xudong Kang
4. Jon Atli Benediktsson
5. Devis Tuia
6. M. Dalla Mura
7. Zebin Wu
8. Guangrui Zhao

Citations in the corpus

• Deep Learning for Hyperspectral Image Classification: An Overview (2019, unknown): “[13], [14], independent component analysis (ICA) [15], and linear discriminant analysis (LDA) [16].”
• Automatic Feature Learning for Spatio-Spectral Image Classification With Sparse SVM (2014, result): “On the contrary, many features corresponding to higher frequencies (lower variance) are retained, suggesting that very useful information is still present in the small-eigenvalue spectrum part of the PCA components, as observed in the previous literature [47], [48].”
• Spectral–Spatial Feature Tokenization Transformer for Hyperspectral Image Classification (2022, methodology): “In addition, several methods for dimension reduction and spectral information extraction have also been developed, including the principal component analysis (PCA) [22], [23], the independent component analysis (ICA) [24], and the linear discriminant analysis (LDA) [25]–[27].”
• Spectral–Spatial Hyperspectral Image Classification via Multiscale Adaptive Sparse Representation (2014, background): “…have focused on effective feature extraction or feature reduction approaches, including spectral-only based (e.g., principal components analysis [18], linear discriminative analysis [19], and clonal selection for feature selection [20]) and spectral–spatial based (e.g., local Fisher’s…”
• Manifold-Learning-Based Feature Extraction for Classification of Hyperspectral Data: A Review of Advances in Manifold Learning (2014, unknown): “Under the assumption of homoscedastic Gaussian class-conditional distributions, LDA is optimized for classification tasks but does not perform well when the data are heteroscedastic Gaussian, and it can fail for non-Gaussian data.”
• Classification of Hyperspectral Images by Exploiting Spectral–Spatial Information of Superpixel via Multiple Kernels (2015, methodology): “reduction techniques, such as the principle component analysis [17], clonal selection”