Mixed Pixel Classification on Hyperspectral Image Using Imbalanced Learning and Hyperparameter Tuning Methods

Purwadi Purwadi - Universitas AMIKOM Purwokerto, Purwokerto, Indonesia
Nor Abu - Universiti Teknikal Malaysia Melaka (UTeM)
Othman Mohd - Universiti Teknikal Malaysia Melaka (UTeM)
Bagus Kusuma - Universitas AMIKOM Purwokerto, Purwokerto, Indonesia


Citation Format:



DOI: http://dx.doi.org/10.30630/joiv.7.3.1758

Abstract


Hyperspectral image technology in land classification is a distinct advantage compared to ordinary RGB and multispectral images. This technology has a wide spectrum of electromagnetic waves, which can be more detailed than other types of imagery. Therefore, with its hyperspectral advantages, the characteristics of an object should have a high probability of being recognized and distinguished. However, because of the large data, it becomes a challenge to lighten the computational burden. Hyperspectral has a huge phenomenon that makes computations heavy compared to other types of images because this image is 3D. The problem faced in hyperspectral image classification is the high computational load, especially if the spatial resolution of the image also has mixed pixel problems. This research uses EO-1 satellite imagery with a spatial resolution of 30 meters and a mixed pixel problem. This study uses a classification method to lighten the computational burden and simultaneously increase the value of classification accuracy. The method used is satellite image pre-processing, including geometric correction and image enhancement using FLAASH while the corrections are geometric correction and atmospheric correction. Then to lighten the computational burden, the steps carried out are using the Slab and PCA method. After obtaining the characteristics, they are entered into a guided learning model using a support vector machine (SVM) for the five-class or multiclass classification. Moreover, the imbalance learning method is proven to produce increased accuracy. The best results were achieved by the ADASYN method with an accuracy of 96.58%, while the computational time became faster with the feature extraction method.


Keywords


Mixed pixel; imbalanced learning; hyperparameter tuning, machine learning; hyperspectral classification

Full Text:

PDF

References


S. Bera and V. K. Shrivastava, “Analysis of various optimizers on deep convolutional neural network model in the application of hyperspectral remote sensing image classification,†Int J Remote Sens, vol. 41, no. 7, pp. 2664–2683, Apr. 2020, doi: 10.1080/01431161.2019.1694725.

L. Su, Y. Sui, and Y. Yuan, “An Unmixing-Based Multi-Attention GAN for Unsupervised Hyperspectral and Multispectral Image Fusion,†Remote Sens (Basel), vol. 15, no. 4, Feb. 2023, doi: 10.3390/rs15040936.

K. Zhang et al., “Panchromatic and multispectral image fusion for remote sensing and earth observation: Concepts, taxonomy, literature review, evaluation methodologies and challenges ahead,†Information Fusion, vol. 93. Elsevier B.V., pp. 227–242, May 01, 2023. doi: 10.1016/j.inffus.2022.12.026.

Purwadi, N. Suryana, and N. A. Abu, “An understanding rice hyperspectral remote sensing imagery classification framework,†J Theor Appl Inf Technol, vol. 97, no. 3, pp. 879–896, 2019.

Purwadi, N. Suryana, N. A. Abu, and B. A. Kusuma, “A comprehensive review: Classification techniques on hyperspectral remote sensing,†International Journal of Advanced Trends in Computer Science and Engineering, vol. 8, no. 1.5 Special Issue, pp. 156–164, 2019, doi: 10.30534/ijatcse/2019/3181.52019.

N. Gupta and V. K. Panchal, “Artificial intelligence for mixed pixel resolution,†International Geoscience and Remote Sensing Symposium (IGARSS), pp. 2801–2804, 2011, doi: 10.1109/IGARSS.2011.6049796.

B. Fang, Y. Bai, and Y. Li, “Combining spectral unmixing and 3D/2D dense networks with early-exiting strategy for hyperspectral image classification,†Remote Sens (Basel), vol. 12, no. 5, Mar. 2020, doi: 10.3390/rs12050779.

L. Li, T. Xu, and Y. Chen, “Improved urban flooding mapping from remote sensing images using generalized regression neural network-based super-resolution algorithm,†Remote Sens (Basel), vol. 8, no. 8, 2016, doi: 10.3390/rs8080625.

C. Zhang, Q. Wang, H. Xie, Y. Ge, and P. M. Atkinson, “Spatio-temporal subpixel mapping with cloudy images,†Science of Remote Sensing, vol. 6, p. 100068, Dec. 2022, doi: 10.1016/j.srs.2022.100068.

S. Wu et al., “Influence of reconstruction scale, spatial resolution and pixel spatial relationships on the sub-pixel mapping accuracy of a double-calculated spatial attraction model,†Remote Sens Environ, vol. 210, pp. 345–361, Jun. 2018, doi: 10.1016/j.rse.2018.03.015.

K. Mahata, R. Das, S. Das, and A. Sarkar, “Automatic Mixed Pixel Detection using a new Hybrid Cellular Automata Approach on Satellite Image,†in 2017 1st International Conference on Electronics, Materials Engineering and Nano-Technology (IEMENTech), 2017.

X. Meng, H. Shen, H. Li, Q. Yuan, H. Zhang, and L. Zhang, “IMPROVING THE SPATIAL RESOLUTION OF HYPERSPECTRAL IMAGE USING PANCHROMATIC AND MULTISPECTRAL IMAGES : AN,†in Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), 2015 7th Workshop on, 2015, pp. 1–4.

A. Hryvachevskyi, I. Prudyus, L. Lazko, and S. Fabirovskyy, “Improvement of segmentation quality of multispectral images by increasing resolution,†Proceedings of 2016 17th International Conference Computational Problems of Electrical Engineering, CPEE 2016, pp. 2–5, 2016, doi: 10.1109/CPEE.2016.7738750.

S. Patteti, B. Samanta, and D. Chakravarty, “Design of a feature-tuned ANN model based on bulk rock-derived mineral spectra for endmember classification of a hyperspectral image from an iron ore deposit,†Int J Remote Sens, vol. 36, no. 8, pp. 2037–2062, Apr. 2015, doi: 10.1080/01431161.2015.1031920.

W. Feng, W. Huang, H. Ye, and L. Zhao, “Synthetic Minority Over-Sampling Technique Based Rotation Forest For The Classification Of Unbalanced Hyperspectral Data,†in IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, IEEE, 2018, pp. 2651–2654.

D. Datta et al., “A Hybrid Classification of Imbalanced Hyperspectral Images Using ADASYN and Enhanced Deep Subsampled Multi-Grained Cascaded Forest,†Remote Sens (Basel), vol. 14, no. 19, Oct. 2022, doi: 10.3390/rs14194853.

W. Feng, W. Huang, and W. Bao, “Imbalanced Hyperspectral Image Classification with an Adaptive Ensemble Method Based on SMOTE and Rotation Forest with Differentiated Sampling Rates,†IEEE Geoscience and Remote Sensing Letters, vol. 16, no. 12, pp. 1879–1883, Dec. 2019, doi: 10.1109/LGRS.2019.2913387.

J. F. R. Rochac, N. Zhang, L. Thompson, and T. Oladunni, “A Data Augmentation-Assisted Deep Learning Model for High Dimensional and Highly Imbalanced Hyperspectral Imaging Data,†in 9th International Conference on Information Science and Technology : ICIST2019 final program, 2019, pp. 362–367.

Y. Quan, X. Zhong, W. Feng, J. C. W. Chan, Q. Li, and M. Xing, “Smote-based weighted deep rotation forest for the imbalanced hyperspectral data classification,†Remote Sens (Basel), vol. 13, no. 3, pp. 1–25, Jan. 2021, doi: 10.3390/rs13030464.

A. Özdemir, K. Polat, and A. Alhudhaif, “Classification of imbalanced hyperspectral images using SMOTE-based deep learning methods,†Expert Syst Appl, vol. 178, Sep. 2021, doi: 10.1016/j.eswa.2021.114986.

L. N. Eeti and K. M. Buddhiraju, “Classification Of Hyperspectral Remote Sensing Images By An Ensemble Of Support Vector Machines Under Imbalanced Data,†in IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, IEEE, 2018, pp. 2659–2661.

D. Datta, P. K. Mallick, J. Shafi, J. Choi, and M. F. Ijaz, “Computational Intelligence for Observation and Monitoring: A Case Study of Imbalanced Hyperspectral Image Data Classification,†Comput Intell Neurosci, vol. 2022, 2022, doi: 10.1155/2022/8735201.

M. Deák et al., “Heterogeneous forest classification by creating mixed vegetation classes using EO-1 Hyperion,†Int J Remote Sens, vol. 38, no. 18, pp. 5215–5231, 2017, doi: 10.1080/01431161.2017.1325529.

S. A. Medjahed and M. Ouali, “A new hybrid SSA-TA: Salp Swarm Algorithm with threshold accepting for band selection in hyperspectral images,†Applied Soft Computing Journal, vol. 95, 2020, doi: 10.1016/j.asoc.2020.106534.

M. A. Toksöz and I. Ulusoy, “Hyperspectral Image Classification via Sparse Representation,†IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, pp. 1–12, 2017.

Q. Gao, S. Lim, and X. Jia, “Hyperspectral image classification using convolutional neural networks and multiple feature learning,†Remote Sens (Basel), vol. 10, no. 2, 2018, doi: 10.3390/rs10020299.

M. P. Uddin, M. Al Mamun, and M. A. Hossain, “PCA-based Feature Reduction for Hyperspectral Remote Sensing Image Classification,†IETE Technical Review (Institution of Electronics and Telecommunication Engineers, India), vol. 38, no. 4. Taylor and Francis Ltd., pp. 377–396, 2021. doi: 10.1080/02564602.2020.1740615.

B. Fang, Y. Bai, and Y. Li, “Combining spectral unmixing and 3D/2D dense networks with early-exiting strategy for hyperspectral image classification,†Remote Sens (Basel), vol. 12, no. 5, Mar. 2020, doi: 10.3390/rs12050779.

M. P. Uddin, M. Al Mamun, and M. A. Hossain, “Effective feature extraction through segmentation-based folded-PCA for hyperspectral image classification,†Int J Remote Sens, vol. 40, no. 18, pp. 7190–7220, Sep. 2019, doi: 10.1080/01431161.2019.1601284.

N. Audebert, B. Le Saux, and S. Lefevre, “Deep learning for classification of hyperspectral data: A comparative review,†IEEE Geosci Remote Sens Mag, vol. 7, no. 2, pp. 159–173, Jun. 2019, doi: 10.1109/MGRS.2019.2912563.

H. Liu et al., “Using t-distributed Stochastic Neighbor Embedding (t-SNE) for cluster analysis and spatial zone delineation of groundwater geochemistry data,†J Hydrol (Amst), vol. 597, Jun. 2021, doi: 10.1016/j.jhydrol.2021.126146.

L. Gao, D. Gu, L. Zhuang, J. Ren, D. Yang, and B. Zhang, “Combining t-Distributed Stochastic Neighbor Embedding with Convolutional Neural Networks for Hyperspectral Image Classification,†IEEE Geoscience and Remote Sensing Letters, vol. 17, no. 8, pp. 1368–1372, Aug. 2020, doi: 10.1109/LGRS.2019.2945122.