Band selection and classification of hyperspectral images using Mutual Information: An algorithm based on minimizing the error probability using the inequality of Fano

Abstract

Hyperspectral image is a substitution of more than a hundred images, called bands, of the same region. They are taken at juxtaposed frequencies. The reference image of the region is called Ground Truth map (GT). the problematic is how to find the good bands to classify the pixels of regions; because the bands can be not only redundant, but a source of confusion, and decreasing so the accuracy of classification. Some methods use Mutual Information (MI) and threshold, to select relevant bands without treatement of redundancy; others consider the neighbors having sensibly the same MI with the GT as redundant and so discarded. This is the most inconvenient of this method, because this avoids the advantage of hyperspectral images: some precious information can be discarded. In this paper well make difference between useful and useless redundancy. A band contains useful redundancy if it contributes to decreasing error probability. According to this scheme, we introduce new algorithm using also mutual information, but it retains only the bands minimizing the error probability of classification. To control redundancy, we introduce a complementary threshold. So the good band candidate must contribute to decrease the last error probability augmented by the threshold. This process is a wrapper strategy; it gets high performance of classification accuracy but it is expensive than filter strategy.

Application of Symmetric Uncertainty and Mutual Information to Dimensionality Reduction and Classification of Hyperspectral Images

“…Sarhrouni et al [17] use also a filter strategy based algorithm on MI to select bands. A wrapper strategy based algorithm on MI, Sarhrouni et al [18] is also introduced. By a thresholding, for example with a threshold 0.…”

“…We apply the proposed algorithm on the hyperspectral image AVIRIS 92AV3C [1], 50% of the labelled pixels are randomly chosen and used in training; and the other 50% are used for testing classification [3][17] [18]. The classifier used is the SVM [5][12] [4].…”

Spectral region identification versus individual channel selection in supervised dimensionality reduction of hyperspectral image data

“…approaches. 52,53 We examined a well-known wrapper algorithm for channel selection on the basis of classification accuracy in comparison with a method giving a better classification result in our experiments: SRS-JM combined with MLC. Infact, SRS-JM identifies features based on maximizing class separability, to be applied with a classification algorithm, whereas a wrapper algorithm selects features to maximize classification accuracy.…”

“…43,54 The "one-against-all" strategy converts the problem of k classes (k > 2) into k dual-class problems. The radial basis function 46,52,53 is utilized as the kernel function to map the original feature space into the higher dimensional space. For the kernel width factor (γ) and regularization parameter (C), we applied different values suggested in the literature and chose the ones giving the best classification performance, i.e., γ ¼ 0.1 and C ¼ 2000 proposed by Ref.…”

Different Optimal Band Selection of Hyperspectral Images Using a Continuous Genetic Algorithm

“…One of the most widely used hyperspectral images in the literature (Fu et al, 2006;Jia et al, 2010;Martínez-Usó et al, 2007;Qian et al, 2011;Sarhrouni et al, 2012) was used in this study (Figure 2). This data set was gathered by the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) sensor over the Indian Pines test site in North-western Indiana and consists of 145×145 pixels and 224 spectral reflectance bands in the wavelength range of 0.4-2.5 μm.…”