BAND REDUCTION FOR TARGET DETECTION IN HYPERSPECTRAL IMAGES-Reference-Cited by-同舟云学术

BAND REDUCTION FOR TARGET DETECTION IN HYPERSPECTRAL IMAGES

Published:2020-06-30 Issue:1 Volume:62 Page:53-62
ISSN:1303-6009
Container-title:Communications Faculty of Sciences University of Ankara Series A2-A3 Physical Sciences and Engineering
language:en
Short-container-title:Commun.Fac.Sci.Univ.Ank.Series A2-A3: Phys.Sci. and Eng.

Author:

ŞİMŞEK Murat¹^ORCID,ILGIN Hakki Alparslan¹^ORCID

Affiliation:

1. ANKARA UNIVERSITY

Abstract

Due to the high spectral resolution, hyperspectral images need large data storage and processing time. Indeed, its high dimensional structure requires high computational complexity, especially for target detection. In order to overcome these problems, band reduction methods have been proposed. In this paper, we compare PCA and SNR-based band reduction methods to improve target detection performance in hyperspectral images. Experimental results show that band reduction methods not only reduce processing time, but also increase accuracy rate.

Publisher

Communications Faculty of Sciences University of Ankara Series A2-A3 Physical Sciences and Engineering

Subject

General Medicine

Reference10 articles.

1. Marwaha, R. and Kumar, A., Target Detection Algorithm for Airborne Thermal Hyperspectral Data, ISPRS Technical Commission VIII Symposium, Volume XL-8, 2014.

2. RIT (Rochester Institute of Technology) 2006, Chooke City Hyperspectral Data Set http://dirsapps.cis.rit.edu/blindtest/download. Last access date: 24.07.2019.

3. Rodarmel, C. and Shan, J., Principal Component Analysis for Hyperspectral Image Classification, Surveying and Land Information Science, 62(2), (2002), 115-122.

4. Gonzalez, R. and Woods, R., Digital Image Processing, Reading, Massachusetts, Addison-Wesley Publishing Company. pp.148-56, 1993.

5. Mallapragada, S., Wong, M. and Hung, C., Dimensionality Reduction of Hyperspectral Images for Classification, Ninth International Conference on Information, Tokyo, Japan, 2018.