Pan-sharpening with a Bayesian nonparametric dictionary learning model

Xinghao Ding, Yiyong Jiang, Yue Huang, John Paisley
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:176-184, 2014.

Abstract

Pan-sharpening, a method for constructing high resolution images from low resolution observations, has recently been explored from the perspective of compressed sensing and sparse representation theory. We present a new pan-sharpening algorithm that uses a Bayesian nonparametric dictionary learning model to give an underlying sparse representation for image reconstruction. In contrast to existing dictionary learning methods, the proposed method infers parameters such as dictionary size, patch sparsity and noise variances. In addition, our regularization includes image constraints such as a total variation penalization term and a new gradient penalization on the reconstructed PAN image. Our method does not require high resolution multiband images for dictionary learning, which are unavailable in practice, but rather the dictionary is learned directly on the reconstructed image as part of the inversion process. We present experiments on several images to validate our method and compare with several other well-known approaches.

Cite this Paper

BibTeX
@InProceedings{pmlr-v33-ding14b, title = {{Pan-sharpening with a Bayesian nonparametric dictionary learning model}}, author = {Ding, Xinghao and Jiang, Yiyong and Huang, Yue and Paisley, John}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {176--184}, year = {2014}, editor = {Kaski, Samuel and Corander, Jukka}, volume = {33}, series = {Proceedings of Machine Learning Research}, address = {Reykjavik, Iceland}, month = {22--25 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v33/ding14b.pdf}, url = {https://proceedings.mlr.press/v33/ding14b.html}, abstract = {Pan-sharpening, a method for constructing high resolution images from low resolution observations, has recently been explored from the perspective of compressed sensing and sparse representation theory. We present a new pan-sharpening algorithm that uses a Bayesian nonparametric dictionary learning model to give an underlying sparse representation for image reconstruction. In contrast to existing dictionary learning methods, the proposed method infers parameters such as dictionary size, patch sparsity and noise variances. In addition, our regularization includes image constraints such as a total variation penalization term and a new gradient penalization on the reconstructed PAN image. Our method does not require high resolution multiband images for dictionary learning, which are unavailable in practice, but rather the dictionary is learned directly on the reconstructed image as part of the inversion process. We present experiments on several images to validate our method and compare with several other well-known approaches.} }
Endnote
%0 Conference Paper %T Pan-sharpening with a Bayesian nonparametric dictionary learning model %A Xinghao Ding %A Yiyong Jiang %A Yue Huang %A John Paisley %B Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2014 %E Samuel Kaski %E Jukka Corander %F pmlr-v33-ding14b %I PMLR %P 176--184 %U https://proceedings.mlr.press/v33/ding14b.html %V 33 %X Pan-sharpening, a method for constructing high resolution images from low resolution observations, has recently been explored from the perspective of compressed sensing and sparse representation theory. We present a new pan-sharpening algorithm that uses a Bayesian nonparametric dictionary learning model to give an underlying sparse representation for image reconstruction. In contrast to existing dictionary learning methods, the proposed method infers parameters such as dictionary size, patch sparsity and noise variances. In addition, our regularization includes image constraints such as a total variation penalization term and a new gradient penalization on the reconstructed PAN image. Our method does not require high resolution multiband images for dictionary learning, which are unavailable in practice, but rather the dictionary is learned directly on the reconstructed image as part of the inversion process. We present experiments on several images to validate our method and compare with several other well-known approaches.
RIS
TY - CPAPER TI - Pan-sharpening with a Bayesian nonparametric dictionary learning model AU - Xinghao Ding AU - Yiyong Jiang AU - Yue Huang AU - John Paisley BT - Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics DA - 2014/04/02 ED - Samuel Kaski ED - Jukka Corander ID - pmlr-v33-ding14b PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 176 EP - 184 L1 - http://proceedings.mlr.press/v33/ding14b.pdf UR - https://proceedings.mlr.press/v33/ding14b.html AB - Pan-sharpening, a method for constructing high resolution images from low resolution observations, has recently been explored from the perspective of compressed sensing and sparse representation theory. We present a new pan-sharpening algorithm that uses a Bayesian nonparametric dictionary learning model to give an underlying sparse representation for image reconstruction. In contrast to existing dictionary learning methods, the proposed method infers parameters such as dictionary size, patch sparsity and noise variances. In addition, our regularization includes image constraints such as a total variation penalization term and a new gradient penalization on the reconstructed PAN image. Our method does not require high resolution multiband images for dictionary learning, which are unavailable in practice, but rather the dictionary is learned directly on the reconstructed image as part of the inversion process. We present experiments on several images to validate our method and compare with several other well-known approaches. ER -
APA
Ding, X., Jiang, Y., Huang, Y. & Paisley, J.. (2014). Pan-sharpening with a Bayesian nonparametric dictionary learning model. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:176-184 Available from https://proceedings.mlr.press/v33/ding14b.html.

Related Material