AUTHORS: Koray Kayabol
Download as PDF
ABSTRACT: We propose the sparse multinomial logistic regression (SMLR) model for spectral-spatial classification of hyperspectral images. In the proposed method, the parameters of SMLR are iteratively estimated from logposterior by using Laplace approximation. The proposed update rule provides a faster convergence compared to the state-of the-art methods used for SMLR parameter estimation. The estimated parameters are used for spectralspatial classification of hyperspectral images using a spatial prior. The experimental results on real hyperspectral images show that the classification accuracy of proposed method is also better than those of state-of-the art methods.
KEYWORDS: Sparse multinomial logistic regression, softmax, hyperspectral images, spatial-spectral classification
REFERENCES:
[1] Ian Goodfellow Yoshua Bengio and Aaron Courville. Deep learning. Book in preparation for MIT Press, 2016. 0 10 20 30 40 50 60 70 80 90 100 20 30 40 50 60 70 80 90 100 Number of iterations OA LORSAL CWSMLR APSMLR Figure 1: Iteration number versus overall accuracy (OA) for Salinas data.
[2] J. Besag. On the statistical analysis of dirty pictures. J. R. Stat. Soc. B, 48(3):259302, 1986.
[3] J. Bioucas-Dias and M. Figueiredo. Logistic regression via variable splitting and augmented Lagrangian tools. Technical report, Instituto Superior Tecnico, TULisbon, Lisbon, Portugal, August 2009.
[4] C. M. Bishop. Pattern Recognition and Machine Learning. Springer, 2006.
[5] D. Bohning. Multinomial logistic regression algorithm. Annals of the Inst. of Statistical Math., 44:197–200, 1992.
[6] J. Borges, J. Bioucas-Dias, and A. Marcal. Fast sparse multinomial regression applied to hyperspectral data. In Int. Conf. Image Analysis and Recognition, ICIAR, Povoa de Varzim, Portugal, 2006.
[7] J. Borges, J. Bioucas-Dias, and A. Marcal. Bayesian hyperspectral image segmentation with discriminative class learning. IEEE Trans. Geosci. Remote Sens., 49(6):2151–2164, 2011.
[8] K. Kayabol. Approximate sparse multinomial logistic regression for classification. IEEE Trans. Pattern Anal. Machine Intell., Submitted, 2016.
[9] K. Kayabol and S. Kutluk. Bayesian classifi- cation of hyperspectral images using spatiallyvarying Gaussian mixture model. Digital Signal Processing, 59:106–114, 2016.
[10] K. Kayabol and J. Zerubia. Unsupervised amplitude and texture classification of SAR images with multinomial latent model. IEEE Trans. Image Process., 22(2):561–572, 2013.
[11] B. Krishnapuram, L. Carin, M. A. T. Figueiredo, and A. J. Hartemink. Sparse multinomial logistic regression: Fast algorithms and generalization bounds. IEEE Trans. Pattern Anal. Machine Intell., 27(6):957–968, 2016.
[12] J. Li, J. Bioucas-Dias, and A. Plaza. Hyperspectral image segmentation using a new Bayesian approach with active learning. IEEE Trans. Geosci. Remote Sens., 49(10):3947–3960, 2011.
[13] A. Quarteroni, R. Sacco, and F.Saleri. Numerical Mathematics. Springer-Verlag, New York, 2000.
