An Experimental Study for the Effects of Noise on Hyperspectral Imagery Classification
DOI:
https://doi.org/10.5566/ias.3078Keywords:
edge preserving features, hyperspectral image classification, minimum noise fraction, principal component analyses, support vector machineAbstract
Hyperspectral image (HSI) classification is a very important topic in remote sensing. There are many published methods for HSI classification in the literature. Nevertheless, it is not clear which method is the most robust to noise in HSI data cubes. In this paper, we conduct a systematic study to examine the effects of noise in HSI data cubes on classification methods. We compare ten existing methods for HSI classification when Gaussian white noise (GWN) and shot noise are present in the HSI data cubes. We have figured out which method is the most robust to GWN and shot noise respectively by experimenting on three widely used HSI data cubes. We have also measured the CPU computational time of every method compared in this paper for HSI classification.
References
Chen GY (2021), Multiscale filter-based hyperspectral image classification with PCA and SVM, Electr Eng, 72:40-5.
Chen GY Bui TB Quach KG and Qian SQ (2014), Denoising hyperspectral imagery using principal component analysis and block matching 4D filtering, Can J Remote Sens, 40:60-7.
Chen GY Krzyzak A and Qian SE (2022), Hyperspectral imagery classification with minimum noise fraction, 2D spatial filtering and SVM, Int J Wavelets Multi, 20:2250025.
Chen GY Krzyzak A and Qian S Q (2023), Noise robust hyperspectral image classification with MNF-based edge-preserving features, Image Anal Stereol, 42:93-9.
Chen GY and Qian SQ (2011), Denoising of hyperspectral imagery using principal component analysis and wavelet shrinkage, IEEE T Geosci Remote, 49:973-80.
Chen GY Xie WF Krzyzak A and Qian SE (2021), Hyperspectral image classification via principal component analysis, 2D spatial convolution, and support vector machines, J Appl Remote Sens, 15:032202.
Jiang J, Chen C, Yu Y, Jiang X, and Ma J, Spatial-aware collaborative representation for hyperspectral remote sensing image classification, IEEE Geosci Remote S, vol. 14, no. 3, pp. 404-408, 2017.
Kang X., Li S., Fang L, Li M and Benediktsson J A. (2015), Extended random walker-based classification of hyperspectral images, IEEE T Geosci Remote, 53:144-53.
Kang X, Li C, Li S and Lin H (2018), Classification of hyperspectral images by Gabor filtering based deep network, IEEE J Sel Top Appl, 11:1166-78.
Kang X, Li S and Benediktsson, JA (2014), Spectral-spatial hyperspectral image classification with edge-preserving filtering, IEEE T Geosci Remote, 52:2666-77.
Kang X, Xiang X, Li S and Benediktsson JA (2017), PCA-based edge-preserving features for hyperspectral image classification, IEEE T Geosci Remote, 55:7140-51.
Li S, Hao Q, Kang X and Benediktsson JA (2018) Gaussian pyramid based multiscale feature fusion for hyperspectral image classification, IEEE J Sel Top Appl, 11:3312-24.
Luo GC, Chen GY, Tian L, Qin K and Qian SE (2016), Minimum noise fraction versus principal component analysis as a preprocessing step for hyperspectral imagery denoising, Can J Remote Sens, 42:106-16.
Tu B, Zhang X, Kang X, Zhang G, Wang J and Wu J (2018), Hyperspectral image classification via fusing correlation coefficient and joint sparse representa-tion, IEEE Geosci Remote S, 15:340-4.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Guangyi Chen, Adam Krzyzak, Shen-en Qian
This work is licensed under a Creative Commons Attribution 4.0 International License.
