A Study of A Proposed Suboptimal Selection Strategy Based On Genetic Algorithm And Filters of Mutual Information

A Study of A Proposed Suboptimal Selection Strategy Based On Genetic Algorithm And Filters of Mutual Information

© 2021 by IJETT Journal
Volume-69 Issue-11
Year of Publication : 2021
Authors : Merzouqi Maria, Agouzal Mehdi, Sarhrouni El kebir, Hammouch Ahmed
DOI :  10.14445/22315381/IJETT-V69I11P209

How to Cite?

Merzouqi Maria, Agouzal Mehdi, Sarhrouni El kebir, Hammouch Ahmed, "A Study of A Proposed Suboptimal Selection Strategy Based On Genetic Algorithm And Filters of Mutual Information," International Journal of Engineering Trends and Technology, vol. 69, no. 11, pp. 71-82, 2021. Crossref, https://doi.org/10.14445/22315381/IJETT-V69I11P209

A novel approach to reduce the dimensionality of hyperspectral images which adopts the selection strategy has been proposed in this work. This approach introduces new genetic algorithm hybridization with one of the mutual information filters (CIM, MRLR, MRMR, NMIFS, MIFS_U2, MIFSU, and MIFS), which have been integrated as its fitness function. Also, this work makes a comparative study of the behavior of genetic algorithms with these filters in the selection of bands. The validation of the result is calculated by two classifiers (RBF-SVM and KNN) to judge the behavior of the hybridization introduced using the overall accuracy (OA %). The new approach is performed on hyperspectral datasets (Pavia, Indian Pines, and Salinas). The results obtained reveal that the performance of this hybridization in selecting the minimum of bands (ex: 30 bands of Salinas) exceeds 95% of OA. According to the results, this approach proves that it can select the most relevant bands and control the redundancy positively. In terms of classification accuracy, they have fascinating results in an acceptable processing time. The results confirm that this hybridization can discriminate between the relevant and irrelevant redundancy bands, efficiently reducing the HSI. Especially the first proposal CMI_GA, except that other comparison hybridization could also fight against the large dimension of these images, precisely those which are not controlled by the adjustable parameter (MRLR_GA with OA =93.93% for only 25 bands of Pavia, and OA=93.44 % for 30 bands of Indiana Pine).

Reduction of dimensionality, mutual information filters, genetic algorithm, classification, RBF-SVM, hybridization strategy.

[1] van Ruitenbeek, F. J., Bakker, W. H., van der Werff, H. M., Zegers, T. E., Oosthoek, J. H., Omer, Z. A., ... & van der Meer, F. D., Mapping the wavelength position of deepest absorption features to explore mineral diversity in hyperspectral images. Planetary and space science, 101, (2014) 108-117.
[2] Karakaya, A., & Yüksel, S. E., Target detection in hyperspectral images. In 2016 24th Signal Processing and Communication Application Conference (SIU) (2016) 1501-1504 . IEEE.
[3] Brook, A., Ben-Dor, E., & Richter, R., Fusion of hyperspectral images and LiDAR data for civil engineering structure monitoring. In 2010 2nd Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (2010) 1-5 , IEEE.
[4] Zhang, T., Huang, Y., Reddy, K. N., Yang, P., Zhao, X., & Zhang, J., Using Machine Learning and Hyperspectral Images to Assess Damages to Corn Plant Caused by Glyphosate and to Evaluate Recoverability. Agronomy, 11(3) (2021) 583.
[5] Panasyuk SV, Yang S, Faller DV, Ngo D, Lew RA, Freeman JE, Rogers AE. Medical hyperspectral imaging to facilitate residual tumor identification during surgery. Cancer biology & therapy. 1;6(3) (2007) 439-46.
[6] G.F. Hughes, On the mean accuracy of statistical pattern recognizers. IEEE Transactions on Infirmation Theory, 14(1) (1968) 55-63.
[7] Khodr, J., & Younes, R, Dimensionality reduction on hyperspectral images: A comparative review based on artificial data. In 2011 4th International Congress on Image and Signal Processing , 4 (2011) 1875-1883. IEEE.
[8] Ayesha, S., Hanif, M. K., & Talib, R., Overview and comparative study of dimensionality reduction techniques for high dimensional data. Information Fusion, 59 (2020) 44-58.
[9] Du, Q., & Younan, N. H., Dimensionality reduction and linear discriminant analysis for hyperspectral image classification. In International Conference on Knowledge-Based and Intelligent Information and Engineering Systems Springer, Berlin, Heidelberg. (2008) 392-399.
[10] Du, Q., Modified Fisher`s linear discriminant analysis for hyperspectral imagery. IEEE Geoscience and remote sensing letters, 4(4) (2007) 503-507.
[11] Datta, A., Ghosh, S., & Ghosh, A., PCA, kernel PCA, and dimensionality reduction in hyperspectral images. In Advances in Principal Component Analysis . Springer, Singapore. (2018) 19-46.
[12] Lennon, M., Mercier, G., Mouchot, M. C., & Hubert-Moy, L., Independent component analysis as a tool for dimensionality reduction and the representation of hyperspectral images. In IGARSS 2001. Scanning the Present and Resolving the Future. Proceedings. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No. 01CH37217) 6 (2001) 2893-2895. IEEE.
[13] Dobigeon, N., Tourneret, J. Y., Richard, C., Bermudez, J. C. M., McLaughlin, S., & Hero, A. O., Nonlinear unmixing of hyperspectral images: Models and algorithms. IEEE Signal processing magazine, 31(1) (2013) 82-94.
[14] Cao, F., Yang, Z., Ren, J., Jiang, M., & Ling, W. K., Linear vs. nonlinear extreme learning machine for spectral-spatial classification of hyperspectral images. Sensors, 17(11) (2017) 2603.
[15] Kang, X., Li, S., & Benediktsson, J. A., Feature extraction of hyperspectral images with image fusion and recursive filtering. IEEE Transactions on Geoscience and Remote Sensing, 52(6) (2013) 3742-3752.
[16] Fang, L., He, N., Li, S., Plaza, A. J., & Plaza, J., A new spatial-spectral feature extraction method for hyperspectral images using local covariance matrix representation. IEEE Transactions on Geoscience and Remote Sensing, 56(6) (2018) 3534-3546.
[17] Archibald, R., & Fann, G., Feature selection and classification of hyperspectral images with support vector machines. IEEE Geoscience and remote sensing letters, 4(4) (2007) 674-677.
[18] Serpico, S. B., & Bruzzone, L., A new search algorithm for feature selection in hyperspectral remote sensing images. IEEE Transactions on Geoscience and Remote Sensing, 39(7) (2001) 1360-1367.
[19] Santos, A. C. S., & Pedrini, H., A combination of k-means clustering and entropy filtering for band selection and classification in hyperspectral images. International Journal of Remote Sensing, 37(13) (2016) 3005-3020.
[20] Cao, X., Ji, B., Ji, Y., Wang, L., & Jiao, L., Hyperspectral image classification based on filtering: a comparative study. Journal of Applied Remote Sensing, 11(3) (2017) 035007.
[21] Yao, X., & Zhao, C., Hyperspectral anomaly detection based on the bilateral filter. Infrared Physics & Technology, 92 (2018) 144-153.
[22] Tu, B., Yang, X., Li, N., Ou, X., & He, W., Hyperspectral image classification via superpixel correlation coefficient representation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(11) (2018) 4113-4127.
[23] Sawant, S. S., & Prabukumar, M.,. A survey of band selection techniques for hyperspectral image classification. Journal of Spectral Imaging, 9 (2020).
[24] Guo, B., Gunn, S. R., Damper, R. I., & Nelson, J. D., Band selection for hyperspectral image classification using mutual information. IEEE Geoscience and Remote Sensing Letters, 3(4) (2006) 522-526.
[25] Datta, A., Ghosh, S., & Ghosh, A., Wrapper-based feature selection in hyperspectral image data using self-adaptive differential evolution. In 2011 International Conference on Image Information Processing (2011) 1-6. IEEE
[26] Merzouqi, M., Sarhrouni, E., & Hammouch, A., Classification and reduction of hyperspectral images based on the motley method. In Proceedings of the Computational Methods in Systems and Software Springer, Cham. (2019) 155-164.
[27] Sarhrouni, E., Hammouch, A., & Aboutajdine, D., Dimensionality reduction and classification feature using mutual information applied to hyperspectral images: a wrapper strategy algorithm based on minimizing the error probability using the inequality of Fano. arXiv preprint arXiv:1211.0055. (2012).
[28] Loggenberg, K., & Poona, N., A Feature Selection Approach for Terrestrial Hyperspectral Image Analysis. South African Journal of Geomatics, 9(2) (2020) 302-320.
[29] Boggavarapu, L. P. K., & Manoharan, P., A new framework for hyperspectral image classification using Gabor embedded patch-based convolutional neural network. Infrared Physics & Technology, 110 (2020) 103455.
[30] Maria, M., El Kebir, S., & Ahmed, H., Reduction dimensionality of hyperspectral imagery using genetic algorithm and mutual information and normalized mutual information as a fitness function. International Review of Applied Sciences and Engineering, 12(1) (2021) 64-75.
[31] Guo, B., Gunn, S., Damper, B., & Nelson, J., Adaptive band selection for hyperspectral image fusion using mutual information. In 2005 7th International Conference on Information Fusion 1 (2005) 8. IEEE.
[32] Verdú, S., Generalizing the Fano inequality. IEEE Transactions on Information Theory, 40(4) (1994) 1247-1251.
[33] Kwak, N., Feature extraction is based on the direct calculation of mutual information. International Journal of Pattern Recognition and Artificial Intelligence, 21(07) (2007) 1213-1231.
[34] Banit`ouagua, I., Kerroum, M. A., Hammouch, A., & Aboutajdine, D., Band selection by mutual information for hyperspectral image classification. International Journal of Advanced Intelligence Paradigms, 8(1) (2016) 98-118.
[35] Battiti, R., Using mutual information for selecting features in supervised neural net learning. IEEE Transactions on neural networks, 5(4) (1994) 537-550.
[36] Kwak, N., & Choi, C. H., Input feature selection by mutual information based on the Parzen window. IEEE transactions on pattern analysis and machine intelligence, 24(12) (2002) 1667-1671.
[37] [37] Huang, J., Cai, Y., & Xu, X., A wrapper for feature selection based on mutual information. In 18th International Conference on Pattern Recognition (ICPR`06) 2 (2006) 618-621 . IEEE.
[38] Estévez, P. A., Tesmer, M., Perez, C. A., & Zurada, J. M., Normalized mutual information feature selection. IEEE Transactions on neural networks, 20(2) (2009) 189-201.
[39] Lee, S., Park, Y. T., & d’Auriol, B. J. (2012). A novel feature selection method based on normalized mutual information. Applied Intelligence, 37(1), 100-120.
[40] Peng, H., Long, F., & Ding, C., Feature selection is based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Transactions on pattern analysis and machine intelligence, 27(8) (2005) 1226-1238.
[41] Lee, S. M., Islam, M. M., Kim, J., Kim, Y. H., Jeong, I., & Kim, J. M., Filter and Wrapper-based Feature Selection Using Mutual Information for Rolling Elements Bearing Diagnosis. In Proceedings on the International Conference on Artificial Intelligence (ICAI) . The Steering Committee of The World Congress in Computer Science, Computer Engineering and Applied Computing (WorldCom). (2017) 187-190.
[42] Cheng, H., Qin, Z., Qian, W., & Liu, W., Conditional mutual information-based feature selection. In 2008 International Symposium on Knowledge Acquisition and Modeling (2008) 103-107 . IEEE.
[43] Holland, J. H., Adaptation in natural and artificial systems: an introductory analysis of biology, control, and artificial intelligence applications. MIT press. (1992).
[44] Sheikh, R. H., Raghuwanshi, M. M., & Jaiswal, A. N., Genetic algorithm-based clustering: a survey. In First International Conference on Emerging Trends in Engineering and Technology (2008) 314-319 . IEEE.
[45] Nhaila, Hasna, et al. Supervised classification methods applied to airborne hyperspectral images: a comparative study using mutual information. Procedia computer science 148 (2019) 97-106.