Filter selection for multi-spectral image acquisition using the feature vector analysis methods

Ioannis S. Chatzis, Vassilios A. Kappatos, Evangelos S. Dermatas

The selection of filters plays a fundamental role for the setup and the performance of a multispectral image
acquisition system. When the number of available filters is large, the full search method becomes computationally
expensive. In this paper the filter selection problem is defined as a dimensionality reduction process by selecting the filters that contains most of the essential information, using the same criteria as the Principal Component Analysis. The proposed method is also expanded by processing multiple filter estimations to improve the stability and the quality of the results. The filter selection methods were evaluated under the CIE standard illuminants A and D65
using 160 filters and a spectral reflectance database of 1269 curves.

NOTE: TO READ THIS AND OTHER IPROMS 2006 PAPERS, PLEASE REGISTER FOR THE CONFERENCE.

REGISTRATION IS FREE.

CLICK here TO REGISTER.

Attachment	Size
Filter_sel.wmv	3.27 MB

Submitted by soroka on Sun, 02/07/2006 - 12:24pm.
login or register to post comments |

The advantages of filter selection method?

Submitted by LiuH on Mon, 03/07/2006 - 9:21pm.

The proposed method and experiments are very well-explained, but it seems to me that the contributions of the paper are not highlighted, can you specify the advantages of using filter selection method please?

Dear authors, Thanks for

Submitted by LiuH on Wed, 05/07/2006 - 11:57am.

Dear authors,

Thanks for your contribution to IPROMS2006, I am the co-chair of this session and i still look forward to paper presentation. If you see this message, please upload your presentation or send it to me so that i can upload it for you.Thank you.

Contribution of the paper

Submitted by Chatzis on Mon, 10/07/2006 - 9:22am.

The contributions of the paper are:
1) An approach for fast selection of a filter subset given a large set of filters as a part of a multispectral acquisition system, through the use of feature selection methods.
2) The comparison of two appropriate feature selection methods.
3) Evaluation of the correlation of the information thoretic optimization criterion with the spectral reconstruction and colorimetric objectives of the multispectral system.

Filter selection for multi-spectral image acquisition using the feature vector analysis methods

Comment viewing options

User login