Hiperspektral görüntülerde Relief-F algoritması ile band seçimi

Year 2024, Volume: 13 Issue: 3, 766 - 775, 15.07.2024

Mehmet Yılmaz , Ümit Haluk Atasever

https://doi.org/10.28948/ngumuh.1408200

Abstract

Hiperspektral görüntüler, sınıflandırması için detaylı bilgi içermektedirler. Ancak bu veriler yüksek boyut, büyük veri hacmi ve bitişik bantlar arasındaki güçlü korelasyon özellikleri nedeniyle sınıflandırma sonuçlarını olumsuz etkilenmektedir. Uygun bir öznitelik seçim yöntemi ile hiperspektral görüntülerin sınıflandırma etkinliği ve doğruluğu iyileştirilebilir. Bu çalışmada sınıflandırma modelinden bağımsız olması, çoklu bağlantı varsayımını dikkate almaması, gürültü değerlerini işleyebilmesi gibi özellikleri nedeniyle Relief-F öznitelik seçme algoritması tercih edilmiştir. Relief-F algoritmasının uygulama etkisini incelemek için Salinas-A, Indian Pines ve Pavia University veri setleri, deneysel veri olarak kullanılmıştır. Gerçekleştirilen uygulamalar sonrasında band seçimi sonrası Salinas-A, Indian Pines verisetlerinde Destek Vektör Makineleri sınıflandırıcısının daha yüksek performans gösterirken; Rastgele Orman yöntemininin sınıflandırma doğruluğunun büyük oranda korunduğu görülmüştür. Araştırma sonuçları, Relief-F algoritmasının hiperspektral görüntülerde en gerekli özelliklerini belirlemek ve iyi bir sınıflandırma doğruluğu ile bant sayısının %60 - %70 azaltılabileceği göstermektedir.

Keywords

Hiperspektral görüntü sınıflandırma, Öznitelik seçimi, Relief-F

Band selection with Relief-F algorithm in hyperspectral ımages

Abstract

Hyperspectral images contain detailed information for classification. However, these data negatively affect the classification results due to their high size, large data volume and strong correlation between adjacent bands. Classification efficiency and accuracy of hyperspectral images can be improved with an appropriate feature selection method. In this study, the Relief-F feature selection algorithm was preferred due to its features such as being independent of the classification model, not taking into account the assumption of multicollinearity, and being able to process noise values. Salinas-A, Indian Pines and Pavia University datasets were used as experimental data to examine the application effect of the Relief-F algorithm. After the applications, the Support Vector Machine classifier showed higher performance in the Salinas-A and Indian Pines datasets after band selection; It has been observed that the classification accuracy of the Random Forest method is largely preserved. The research results show that the Relief-F algorithm determines the most necessary features in hyperspectral images and the number of bands can be reduced by 60% - 70% with a good classification accuracy.

Keywords

Hyperspectral image classification, feature selection, Relief-F

References

• A. Ghosh, A. Datta and S. Ghosh, Self-adaptive differential evolution for feature selection in hyperspectral image data. Applied Soft Computing, 13 (4), 1969-1977, 2013. https://doi.org/10.1016/j.asoc.2 012.11.042.
• S. Y. Xiang, Z. H. Xu, Y. W. Zhang, Q. Zhang, X. Zhou, H. Yu, B. Li and Y. F. Li, Construction and Application of ReliefF-RFE Feature Selection Algorithm for Hyperspectral Image Classification. Spectroscopy and Spectral Analysis, 42 (10), 3283-3290, 2022.
• B. Wu, C. C. Chen, T. M. Kechadi and L. Y. Sun, A comparative evaluation of filter-based feature selection methods for hyper-spectral band selection. International Journal of Remote Sensing, 34 (22), 7974-7990, 2013. https://doi.org/10.1080/01431161.20 13.827815.
• J. S. Ren, R. X. Wang, G. Liu, R. Y. Feng, Y. N. Wang and W. Wu, Partitioned Relief-F Method for Dimensionality Reduction of Hyperspectral Images. Remote Sensing, 12 (7), 21, 2020. https://doi.org /10.3390/rs12071104.
• T. Lillesand, R. W. Kiefer and J. Chipman, Remote Sensing and Image Interpretation. Wiley, 2015.
• B. Rasti, D. F. Hong, R. L. Hang, P. Ghamisi, X. D. Kang, J. Chanussot and J. A. Benediktsson, Feature Extraction for Hyperspectral Imagery: The Evolution From Shallow to Deep: Overview and Toolbox. Ieee Geoscience and Remote Sensing Magazine, 8 (4), 60-88, 2020. https://doi.org/10.1109/mgrs.2020.2979764.
• R. Jung and M. Ehlers, Comparison of two feature selection methods for the separability analysis of intertidal sediments with spectrometric datasets in the German Wadden Sea. International Journal of Applied Earth Observation and Geoinformation, 52, 175-191, 2016. https://doi.org/10.1016/j.jag.2016.06.009.
• Y. Dong, B. Du, L. Zhang and L. Zhang, Dimensionality Reduction and Classification of Hyperspectral Images Using Ensemble Discriminative Local Metric Learning. IEEE Transactions on Geoscience and Remote Sensing, 55 (5), 2509-2524, 2017. https://doi.org/10.1109/TGRS.2016.2645703.
• X. Zhang, X. Jiang, J. Jiang, Y. Zhang, X. Liu and Z. Cai, Spectral–Spatial and Superpixelwise PCA for Unsupervised Feature Extraction of Hyperspectral Imagery. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-10, 2022. https://doi.org/10.11 09/TGRS.2021.3057701.
• M. R. Islam, A. Siddiqa, M. Ibn Afjal, M. P. Uddin and A. Ulhaq, Hyperspectral Image Classification via Information Theoretic Dimension Reduction. 15 (4), 1147, 2023.
• X. C. Y. Su and F. Liu, A Survey For Study of Feature Selection Based On Mutual Information. 9th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), pp. 1-4, Amsterdam, Netherlands, 2018.
• Z. H. Wang, S. L. Liang, L. Z. Xu, W. Song, D. X. Wang and D. M. Huang, Dimensionality reduction method for hyperspectral image analysis based on rough set theory. European Journal of Remote Sensing, 53 (1), 192-200, 2020. https://doi.org/10.1080/227972 54.2020.1785949.
• M. C. Ye, Y. Q. Xu, C. X. Ji, H. Chen, H. J. Lu and Y. T. Qian, Feature selection for cross-scene hyperspectral image classification using cross-domain ReliefF. International Journal of Wavelets Multiresolution and Information Processing, 17 (5), 17, 2019. https://doi. org/10.1142/s0219691319500395.
• A. Elmaizi, E. Sarhrouni, A. Hammouch and C. Nacir, A new band selection approach based on information theory and support vector machine for hyperspectral images reduction and classification. International Symposium on Networks, Computers and Communications (ISNCC), pp. 1-6, Marrakech, Morocco, 2017. https://doi.org/10.1109/ISNCC.2017. 8072002.
• S. Zhou, J. P. Zhang and B. K. Su, Feature selection and classification based on ant colony algorithm for hyperspectral remote sensing images. 2nd International Congress on Image and Signal Processing, pp. 1-4, Tianjin, China, 2009. https://doi.org/10.1109/CISP.20 09.5304614.
• W. W. Sun and Q. Du, Hyperspectral Band Selection A review. Geoscience and Remote Sensing Magazine, 7 (2), 118-139, 2019. https://doi.org/10.1109/mgrs.2019. 2911100.
• K. Kira, L. A. Rendell, A Practical Approach To Feature-Selection. 9th International Workshop on Machine Learning, pp. 249-256, Aberdeen, Scotland, 1992.
• S. Sevindik, Diskriminant analizi ve bazı alternatif regresyon analizleri. Yüksek Lisans Tezi, Çukurova Üniversitesi Fen Bilimleri Enstitüsü, Türkiye, 2018.
• M. Belgiu and L. Dragut, Random forest in remote sensing: A review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing, 114, 24-31, 2016. https://doi.org/10.1016/j.is prsjprs.2016.01.011.
• O. Kramer, K-Nearest Neighbors. in Dimensionality Reduction with Unsupervised Nearest Neighbors. O. Kramer, Ed. Berlin, Heidelberg: Springer, pp. 13-23, 2013.
• M. Awad and R. Khanna, Support Vector Machines for Classification. in Efficient Learning Machines: Theories, Concepts, and Applications for Engineers and System Designers. M. Awad and R. Khanna, Eds. Apress Berkeley, CA, pp. 39-66, 2015.
• M. Robnik-Sikonja and I. Kononenko, Theoretical and empirical analysis of ReliefF and RReliefF. Machine Learning, 53 (1-2), 23-69, 2003. https://doi.org/10.10 23/a:1025667309714.
• K. Kira and L. A. Rendell, The Feature Selection Problem: Traditional Methods and a New Algorithm. In Proceedings of the 10th AAAI Conference on Artificial Intelligence, pp. 129–134, California, ABD, July 12-16, 1992.
• S. Riyanto, I. S. Sitanggang, T. Djatna and T. D. Atikah, Comparative Analysis using Various Performance Metrics in Imbalanced Data for Multi-class Text Classification. International Journal of Advanced Computer Science and Applications, 14 (6), 1082-1090, 2023.
• L. Cuadros-Rodríguez, E. Pérez-Castaño and C. Ruiz-Samblás, Quality performance metrics in multivariate classification methods for qualitative analysis. TrAC Trends in Analytical Chemistry, 80, 612-624, 2016. https://doi.org/10.1016/j.trac.2016.04.021.
• M. A. Günen, U. H. Atasever, E. Besdok, Analyzing the Contribution of Training Algorithms on Deep Neural Networks for Hyperspectral Image Classification. Photogrammetric Engineering and Remote Sensing 86 (9): 581-588, 2020.