论文信息 - Unsupervised band selection for multispectral images using information theory

Unsupervised band selection for multispectral images using information theory

In this paper, the implication of the relations of information in the case of multispectral images is analyzed. Higher-order mutual information can adopt positive or negative values depending of the correlation among ensembles. Therefore, the existence of negative values reflects higher-order correlations in the conditional information. On the other hand, the extraction of optimal subsets of spectral images is proposed as a maximization of the conditional entropies at same time that the dependent information among images is minimized.

Filiberto Pla | José Martínez Sotoca | Arnoud C. Klaren | F. Pla | J. Sotoca

[1] Thomas M. Cover,et al. Elements of Information Theory , 2005 .

[2] T. Tsujishita,et al. On Triple Mutual Information , 1994 .

[3] Igor Kononenko,et al. Estimating Attributes: Analysis and Extensions of RELIEF , 1994, ECML.

[4] Matsuda,et al. Physical nature of higher-order mutual information: intrinsic correlations and frustration , 2000, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[5] M K Markey,et al. Application of the mutual information criterion for feature selection in computer-aided diagnosis. , 2001, Medical physics.

[6] Max A. Viergever,et al. Mutual-information-based registration of medical images: a survey , 2003, IEEE Transactions on Medical Imaging.

[7] A. Kercek,et al. Real-time classification of polymers with NIR spectral imaging and blob analysis , 2003, Real Time Imaging.

[8] Andrew R. Harvey,et al. Spectral imaging in a snapshot , 2000, SPIE BiOS.