DAE-MLP Based Feature Extraction for Hyperspectral Image Classification of Saint Clair River
Abstract
Hyperspectral remote sensing has emerged as a powerful tool for vegetation classification due to its ability to capture detailed spectral information. This study introduces a novel methodology for vegetation classification using exclusively hyperspectral imagery. The proposed approach comprises atmospheric correction using the FLAASH algorithm, followed by dimensionality reduction
using PCA and segmentation through the ROI selection and the Spectral Angle Mapper (SAM) module. Subsequently, a deep autoencoder is employed for feature extraction, paving the way for classification using the Multi-Layer Perceptron (MLP) algorithm. The effectiveness of this methodology is evaluated using a hyperspectral image of the Saint Clair River, successfully classifying the image into six main classes: water 1, water 2, grass, tree, reed, corn, and an 'unclassified' category encompassing concrete, roads, bricks, wood, and more. Our findings demonstrate the efficacy of this approach in accurately classifying and mapping vegetation in river ecosystems, offering a promising solution in the face of limited hyperspectral datasets.
Keywords
Hyperspectral remote sensing, FLAASH, PCA, SAM, Multi-Layer Perceptron (MLP), Saint Clair RiverReferences
G. Ortac et G. Ozcan, « Comparative study of hyperspectral image classification by
multidimensional Convolutional Neural Network approaches to improve
accuracy », Expert Systems with Applications, vol. 182, p. 115280, nov. 2021, doi:
https://doi.org/10.1016/j.eswa.2021.115280.
M. Zhao, J. Chen, et Z. He, « A Laboratory-Created Dataset With Ground Truth for
Hyperspectral Unmixing Evaluation », IEEE Journal of Selected Topics in Applied
Earth Observations and Remote Sensing, vol. 12, no 7, p. 2170‑2183, juill. 2019,
doi: 10.1109/JSTARS.2019.2905099.
F. Kruse, « Comparison of AVIRIS and Hyperion for hyperspectral mineral
mapping », janv. 2002.
B. Rasti, P. Scheunders, P. Ghamisi, G. Licciardi, et J. Chanussot, « Noise
Reduction in Hyperspectral Imagery: Overview and Application », Remote Sensing,
vol. 10, no 3, Art. no 3, mars 2018, doi: 10.3390/rs10030482.
M. M. Baustian et al., « A one hundred year review of the socioeconomic and
ecological systems of Lake St. Clair, North America », Journal of Great Lakes
Research, vol. 40, no 1, p. 15‑26, mars 2014, doi: 10.1016/j.jglr.2013.11.006.
W. Zhao et S. Du, « Spectral–Spatial Feature Extraction for Hyperspectral Image
Classification: A Dimension Reduction and Deep Learning Approach », IEEE
Transactions on Geoscience and Remote Sensing, vol. 54, no 8, p. 4544‑4554, août
, doi: 10.1109/TGRS.2016.2543748.
X. Ma, H. Wang, et J. Geng, « Spectral–Spatial Classification of Hyperspectral
Image Based on Deep Auto-Encoder », IEEE Journal of Selected Topics in Applied
Earth Observations and Remote Sensing, vol. 9, no 9, p. 4073‑4085, sept. 2016, doi:
1109/JSTARS.2016.2517204.
S. Li, X. Zhu, Y. Liu, et J. Bao, « Adaptive Spatial-Spectral Feature Learning for
Hyperspectral Image Classification », IEEE Access, vol. 7, p. 61534‑61547, 2019,
doi: 10.1109/ACCESS.2019.2916095.
J. Li, J. M. Bioucas-Dias, et A. Plaza, « Semisupervised Hyperspectral Image
Segmentation Using Multinomial Logistic Regression With Active Learning »,
IEEE Transactions on Geoscience and Remote Sensing, vol. 48, no 11, p. 4085‑4098,
nov. 2010, doi: 10.1109/TGRS.2010.2060550.
L. Breiman, « Random Forests », Machine Learning, vol. 45, no 1, p. 5‑32, oct. 2001,
doi: 10.1023/A:1010933404324.
C. Bo, H. Lu, et D. Wang, « Spectral-spatial K-Nearest Neighbor approach for
hyperspectral image classification », Multimed Tools Appl, vol. 77, no 9, p.
‑10436, mai 2018, doi: https://doi.org/10.1007/s11042-017-4403-9.
Z. He, D. Lin, T. Lau, et M. Wu, « Gradient Boosting Machine: A Survey ». arXiv,
août 2019. doi: 10.48550/arXiv.1908.06951.
P. Mahesh et F. Giles M, « Feature Selection for Classification of Hyperspectral
Data by SVM », IEEE Transactions on Geoscience and Remote Sensing, vol. 48, no
, p. 2297‑2307, févr. 2010, doi: 10.1109/TGRS.2009.2039484.
K. Deng, H. Zhao, N. Li, et W. Wei, « Identification of minerals in hyperspectral
imagery based on the attenuation spectral absorption index vector using a multilayer
perceptron », Remote Sensing Letters, vol. 12, no 5, p. 449‑458, mai 2021, doi:
1080/2150704X.2021.1903612.
Z. Lin, Y. Chen, X. Zhao, et G. Wang, « Spectral-spatial classification of
hyperspectral image using autoencoders », in 2013 9th International Conference on
Information, Communications & Signal Processing, déc. 2013, p. 1‑5. doi:
1109/ICICS.2013.6782778.
J. Zabalza et al., « Novel segmented stacked autoencoder for effective
dimensionality reduction and feature extraction in hyperspectral imaging »,
Neurocomputing, vol. 185, p. 1‑10, avr. 2016, doi:
https://doi.org/10.1016/j.neucom.2015.11.044.
M. K. Tripathi et H. Govil, « Evaluation of AVIRIS-NG hyperspectral images for
mineral identification and mapping », Heliyon, vol. 5, no 11, p. e02931, nov. 2019,
doi: 10.1016/j.heliyon.2019.e02931.
R. F. Kokaly, et al., « USGS digital spectral library splib06a: U.S. Geological
Survey Data Series 231. », Data Series, 2007.
G. W. Felde et al., « Analysis of Hyperion data with the FLAASH atmospheric
correction algorithm », in IGARSS 2003. 2003 IEEE International Geoscience and
Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477), juill. 2003,
p. 90‑92 vol.1. doi: 10.1109/IGARSS.2003.1293688.
S. L. Chattoraj, R. U. Sharma, C. Kumar, P. K. Champati ray, et V. Sengar,
« Identification and characterization of hydrothermally altered minerals using
surface and space-based reflectance spectroscopy, in parts of south-eastern
Rajasthan, India », SN Appl. Sci., vol. 2, no 4, p. 531, mars 2020, doi:
1007/s42452-020-2225-2.
R. Nisha, R. M. Venkata, et S. Tejpal, « Evaluation of atmospheric corrections on
hyperspectral data with special reference to mineral mapping », Geoscience
Frontiers, vol. 8, no 4, p. 797‑808, juill. 2017, doi: 10.1016/j.gsf.2016.06.004.
H. Li, J. Cui, X. Zhang, Y. Han, et L. Cao, « Dimensionality Reduction and
Classification of Hyperspectral Remote Sensing Image Feature Extraction »,
Remote Sensing, vol. 14, no 18, Art. no 18, janv. 2022, doi: 10.3390/rs14184579.
C. Rodarmel et J. Shan, « Principal Component Analysis for Hyperspectral Image
Classification », Surv Land inf Syst, vol. 62, janv. 2002.
F. A. Kruse et al., « The spectral image processing system (SIPS)—interactive
visualization and analysis of imaging spectrometer data », Remote Sensing of
Environment, vol. 44, no 2, p. 145‑163, mai 1993, doi: https://doi.org/10.1016/0034-
(93)90013-N.
S. Roland, « Athlete Identification and Personel Profile Creation based on
Movement Sensor Data », Master Thesis, Friedrich-Alexander-Universität,
Erlangen, Allemagne, 2023.
Y. Bengio, P. Lamblin, D. Popovici, et H. Larochelle, « Greedy Layer-Wise
Training of Deep Networks », in Advances in Neural Information Processing
Systems, MIT Press, 2006.
joseph, « Pytorch CNN Autoencoder for Image Compression - reason.town ». [En
ligne]. Disponible sur: https://reason.town/pytorch-cnn-autoencoder/
D. Bank, N. Koenigstein, et R. Giryes, « Autoencoders ». arXiv, 3 avril 2021. doi:
48550/arXiv.2003.05991.
A. Das, D. K. S. A. Viji, et L. Sebastian, « A Survey on Image and Video Super
Resolution with Deep Learning Models », International Journal of Engineering
Research & Technology, vol. 9, no 13, août 2021, doi:
17577/IJERTCONV9IS13007.
P. M. Atkinson et A. R. L. Tatnall, « Introduction Neural networks in remote
sensing », International Journal of Remote Sensing, vol. 18, p. 699‑709, mars 1997,
doi: 10.1080/014311697218700.
F. Pacifici, F. Del Frate, C. Solimini, et W. J. Emery, « Neural Networks for Land
Cover Applications », in Computational Intelligence for Remote Sensing, M. Graña
et R. J. Duro, Éd., in Studies in Computational Intelligence. , Berlin, Heidelberg:
Springer, 2008, p. 267‑293. doi: 10.1007/978-3-540-79353-3_11.
A. J. Alberg, J. W. Park, B. W. Hager, M. V. Brock, et M. Diener-West, « The Use
of “Overall Accuracy” to Evaluate the Validity of Screening or Diagnostic Tests »,
Journal of General Internal Medicine, vol. 19, no 5p1, p. 460‑465, 2004, doi:
1111/j.1525-1497.2004.30091.x.
J. C. Leszek, K. Ryszard, O. Arkadiusz, W. Konrad, M. B. Alfred, et P. Nicolai,
Computer Vision and Graphics, ICCVG 2018. in International Conference, ICCVG.
Warsaw, Poland: Springer International Publishing, 2018.
C. Zhang et al., « A hybrid MLP-CNN classifier for very fine resolution remotely
sensed image classification », ISPRS Journal of Photogrammetry and Remote
Sensing, vol. 140, p. 133‑144, juin 2018, doi: 10.1016/j.isprsjprs.2017.07.014.
C. Xing, L. Ma, et X. Yang, « Stacked Denoise Autoencoder Based Feature
Extraction and Classification for Hyperspectral Images », Journal of Sensors, vol.
, p. e3632943, nov. 2015, doi: 10.1155/2016/3632943.
Published
How to Cite
Downloads
Copyright (c) 2025 Youcef Attallah
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
