Improved Hyperspectral Imagery Classification with Adaptive Denoising.

Saved in:

Bibliographic Details
Title:	Improved Hyperspectral Imagery Classification with Adaptive Denoising.
Authors:	Chen, Guang Yi¹ guang_c@cse.concordia.ca, Sun, Wei² wei.sun@concordia.ca, Qian, Shen-En³ shen-en.qian@asc-csa.gc.ca
Source:	Engineering Letters. Jul2026, Vol. 34 Issue 7, p2710-2715. 6p.
Subjects:	Image denoising, Principal components analysis, Hyperspectral imaging systems, Feature extraction, Signal denoising, Image recognition (Computer vision), Noise control
Abstract:	In today's hyperspectral imagery (HSI) data cubes, there always exists a small amount of noise no matter how good the imaging sensors are. As a result, there is a need to reduce this amount of noise to improve classification accuracy. In this paper, we propose to improve hyperspectral imagery classification by introducing denoising adaptively. We do not add noise to the HSI data cubes but reduce noise from them by means of principal component analysis (PCA) preprocessing. We reduce noise for the noisy PCA output bands by means of video block matching and 3D filtering and then perform inverse PCA to obtain denoised data cubes. We combine this denoising technique with segmented PCA and the Gaussian pyramid decomposition-based multiscale feature fusion. Experiments demonstrate that the classification accuracies for the proposed method improve considerably for three well-known HSI data cubes. [ABSTRACT FROM AUTHOR]
	Copyright of Engineering Letters is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	In today's hyperspectral imagery (HSI) data cubes, there always exists a small amount of noise no matter how good the imaging sensors are. As a result, there is a need to reduce this amount of noise to improve classification accuracy. In this paper, we propose to improve hyperspectral imagery classification by introducing denoising adaptively. We do not add noise to the HSI data cubes but reduce noise from them by means of principal component analysis (PCA) preprocessing. We reduce noise for the noisy PCA output bands by means of video block matching and 3D filtering and then perform inverse PCA to obtain denoised data cubes. We combine this denoising technique with segmented PCA and the Gaussian pyramid decomposition-based multiscale feature fusion. Experiments demonstrate that the classification accuracies for the proposed method improve considerably for three well-known HSI data cubes. [ABSTRACT FROM AUTHOR]
ISSN:	1816093X