RECONSTRUCTION OF HYPERSPECTRAL IMAGE BASED ON REGRESSION ANALYSIS - Optimum Regression Model and Channel Selection

Yuji Sakatoku, Jay Arre Toque, Ari Ide-Ektessabi

2009

Abstract

The purpose of this study is to develop an efficient appraoch for producing hyperspectral images by using reconstructed spectral reflectance from multispectral images. In this study, an indirect reconstruction based on regression analysis was employed because of its stability to noise and its practicality. In this approach however, the regression model selection and channel selection when acquiring the multispectral images play important roles, which consequently affects the efficiency and accuracy of reconstruction. The optimum regression model and channel selection were investigated using the Akaike information criterion (AIC). By comparing the model based on the AIC model based on the pseudoinverse method (the pseudinverse method is a widely used reconstruction technique), RMSE could be reduced by fifty percent. In addition, it was shown that AIC-based model has good stability to noise.

References

  1. Hardeberg, J. Y., 2003, Filter Selection for Multispectral Color Image Acquisition, PICS Conference, VA 2003, pp.177-181
  2. Ukishima M., Nakaguchi T., Kato K., Fukuchi Y., Tsumura N., Matsumoto K., Yanagawa N., Morita H., Miyake Y., 2005, An Evaluation of Sharpness in Different Image of an Internal Organ, The 13th Color Imaging Conference, (CIC13), pp.261-263
  3. Carcagni P., Della Patria A., Fontana R., Greco M., Mastroianni M., Materazzi M., Pampaloni E., Pezzati L., 2006, Multispectral imaging of paintings by optical scanning, Optics and Lasers in Engineering 45, pp.360-367
  4. Burns P. D. and Berns R. S., 1996, Analysis of multispectral image capture, Proceedings of the Fourth Color Imaging Conference, pp.19-22
  5. Farrell J. E., Cupitt J., Saunders D., and Wandell B. A., 1999, Estimating spectral reflectances of digital images of art, International Symposium of Multispectral Imaging and Color Reproduction for Digital Archives, pp. 58-64
  6. Shimano N., Terai K., Hironaga M., 2007, Recovery of spectral reflectances of objects being imaged by multispectral camera, Optical Society of America, 24, pp.3211-3219
  7. Cotte P., Dupraz D., 2006, Spectral imaging of Leonardo Da Vinci's Mona Lisa: A true color smile without the influence of aged varnish, CGIV Final Program and Proceedings
  8. Zhao Y., Taplin L. A., Nezamabadi. M, Bern R. S. , 2005, Using the matrix R method for spectral image archives, Proceedings of the 10th Congress of the International Colour Association, pp.469-472
Download


Paper Citation


in Harvard Style

Sakatoku Y., Toque J. and Ide-Ektessabi A. (2009). RECONSTRUCTION OF HYPERSPECTRAL IMAGE BASED ON REGRESSION ANALYSIS - Optimum Regression Model and Channel Selection . In Proceedings of the First International Conference on Computer Imaging Theory and Applications - Volume 1: IMAGAPP, (VISIGRAPP 2009) ISBN 978-989-8111-68-5, pages 50-55. DOI: 10.5220/0001791800500055


in Bibtex Style

@conference{imagapp09,
author={Yuji Sakatoku and Jay Arre Toque and Ari Ide-Ektessabi},
title={RECONSTRUCTION OF HYPERSPECTRAL IMAGE BASED ON REGRESSION ANALYSIS - Optimum Regression Model and Channel Selection },
booktitle={Proceedings of the First International Conference on Computer Imaging Theory and Applications - Volume 1: IMAGAPP, (VISIGRAPP 2009)},
year={2009},
pages={50-55},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001791800500055},
isbn={978-989-8111-68-5},
}


in EndNote Style

TY - CONF
JO - Proceedings of the First International Conference on Computer Imaging Theory and Applications - Volume 1: IMAGAPP, (VISIGRAPP 2009)
TI - RECONSTRUCTION OF HYPERSPECTRAL IMAGE BASED ON REGRESSION ANALYSIS - Optimum Regression Model and Channel Selection
SN - 978-989-8111-68-5
AU - Sakatoku Y.
AU - Toque J.
AU - Ide-Ektessabi A.
PY - 2009
SP - 50
EP - 55
DO - 10.5220/0001791800500055