Research Article
BibTex RIS Cite

On the Use of Multi-Objective Metaheuristic Optimization Algorithms to Increase the Classification Performance of Images Produced from Hybrid Pansharpening Methods

Year 2021, , 1 - 10, 13.03.2021
https://doi.org/10.48123/rsgis.838767

Abstract

Image classification applications require images of high colour quality. However, it is not always possible to obtain such images due to the technical constraints in the sensors and cost problems. Hence, this study, for the first time in the literature, used the multi-objective metaheuristic optimization algorithm Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to improve the performance of the Intensity-Hue-Saturation Discrete Wavelet Transform (IHS-DWT) pansharpening method. Five different pansharpened images with colour quality ranging from best to worst were produced with the proposed method. The input Multispectral (MS) image and all pansharpened images were classified with the Artificial Neural Network (ANN) classifier. The classification results revealed that all pansharpened images produced by the proposed method increased the overall classification accuracy to a certain extent. It was also concluded that the two images with the best colour qualities increased the overall classification accuracy by approximately 24 %. It can also be concluded that the proposed method is able to provide images for applications that require a high classification accuracy.

References

  • Abd El-Samie, F. E., Hadhoud, M. M., & El-Khamy, S. E. (2012). Image Super-Resolution and Applications. CRC press.
  • Deb, K., Pratap, A., Agarwal, S., & Meyarivan, T. A. M. T. (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation, 6(2), 182-197. doi: 10.1109/4235.996017.
  • Garzelli, A., & Nencini, F. (2006a). Fusion of Panchromatic and Multispectral Images by Genetic Algorithms. IEEE International Conference on Geoscience and Remote Sensing Symposium (IGARSS 2006) (3810-3813). Denver, CO, USA. doi: 10.1109/IGARSS.2006.976.
  • Garzelli, A., & Nencini, F. (2006b). PAN-sharpening of Very High Resolution Multispectral Images Using Genetic Algorithms. International Journal of Remote Sensing, 27(15), 3273-3292. doi: 10.1080/01431160600554991.
  • Ghassemian, H. (2016). A review of remote sensing image fusion methods. Information Fusion, 32, 75-89. doi: 10.1016/j.inffus.2016.03.003.
  • Gogineni, R., & Chaturvedi, A. (2018). Sparsity inspired pan-sharpening technique using multi-scale learned dictionary. ISPRS Journal of Photogrammetry and Remote Sensing, 146, 360-372. doi: 10.1016/j.isprsjprs.2018.10.009.
  • Gonzalez, R. C., & Woods, R.E. 2007. Digital Image Processing. 3rd Edition, Pearson.
  • Laben, C. A., & Brower, B. V. (2000). U.S. Patent No. 6,011,875. Washington, DC: U.S. Patent and Trademark Office.
  • Serifoglu Yilmaz, C., Yilmaz, V., & Güngör, O. (2020). On the use of the SOS metaheuristic algorithm in hybrid image fusion methods to achieve optimum spectral fidelity. International Journal of Remote Sensing, 41(10), 3993-4021. doi: 10.1080/01431161.2019.1711244.
  • Serifoglu Yilmaz, C., Yilmaz, V., Gungor, O., & Shan, J. (2019). Metaheuristic pansharpening based on symbiotic organisms search optimization. ISPRS Journal of Photogrammetry and Remote Sensing, 158, 167-187. doi: 10.1016/j.isprsjprs.2019.10.014.
  • Tso, B., & Mather, P. 2009. Classification Methods for Remotely Sensed Data. 2nd Edition, CRC Press.
  • Wald, L. 2000. Quality of high resolution synthesized images: Is there a simple criterion? 3rd Conference: Fusion of Earth Data: Merging Point Measurements, Raster Maps and Remotely Sensed Images (99-103). Sophia Antipolis, France.
  • Yılmaz, V. (2020a). Metasezgisel Guguk Kuşu Arama Algoritması ile Görüntü Kaynaştırma. Türk Uzaktan Algılama ve CBS Dergisi, 1(1), 1-12.
  • Yilmaz, V. (2020b). A Non‐Dominated Sorting Genetic Algorithm‐II‐based approach to optimize the spectral and spatial quality of component substitution‐based pansharpened images. Concurrency and Computation: Practice and Experience, e6030.
  • Yilmaz, V., & Gungor, O. (2016). Determining the optimum image fusion method for better interpretation of the surface of the Earth. Norsk Geografisk Tidsskrift-Norwegian Journal of Geography, 70(2), 69-81. doi: 10.1080/00291951.2015.1126761.
  • Yilmaz, V., Serifoglu Yilmaz, C., Güngör, O., & Shan, J. (2020). A genetic algorithm solution to the gram-schmidt image fusion. International Journal of Remote Sensing, 41(4), 1458-1485. doi: 10.1080/01431161.2019.1667553.
  • Zhou, J., Civco, D. L., & Silander, J. A. (1998). A wavelet transform method to merge Landsat TM and SPOT panchromatic data. International Journal of Remote Sensing, 19(4), 743-757. doi: 10.1080/014311698215973.

Çok Amaçlı Metasezgisel Optimizasyon Algoritmaları ile Hibrit Pan-Keskinleştirme Yöntemlerinden Üretilen Görüntülerin Sınıflandırma Performanslarının Arttırılmasına Yönelik Bir Araştırma

Year 2021, , 1 - 10, 13.03.2021
https://doi.org/10.48123/rsgis.838767

Abstract

Görüntü sınıflandırma uygulamaları yüksek renk kalitesine sahip olan görüntülere ihtiyaç duymaktadır. Ancak, gerek algılayıcılardaki teknik kısıtlamalar, gerekse de maliyet problemlerinden dolayı bu tarz görüntüleri elde etmek her zaman mümkün olmamaktadır. Bu nedenle, bu çalışmada, literatürde ilk defa, çok amaçlı bir metasezgisel optimizasyon algoritması olan Non-Dominated Sorting Genetic Algorithm II (NSGA-II) algoritması, hibrit bir pan-keskinleştirme yöntemi olan Intensity-Hue-Saturation Discrete Wavelet Transform (IHS-DWT) yönteminin performansının iyileştirilmesi amacıyla kullanılmıştır. Önerilen yöntem ile renk kalitesi en iyiden en kötüye değişen beş farklı pan-keskinleştirilmiş görüntü üretilmiştir. Pan-keskinleştirmede kullanılan girdi Çok Bantlı (ÇB) görüntü ile üretilen bütün pan-keskinleştirilmiş görüntüler Yapay Sinir Ağları (YSA) yöntemine göre sınıflandırılmıştır. Sınıflandırma sonuçları, önerilen yöntem ile üretilen bütün pan-keskinleştirilmiş görüntülerin toplam sınıflandırma doğruluğunu belli oranda arttırdığını göstermektedir. Ayrıca, önerilen yöntem ile üretilen en iyi renk kalitesine sahip olan iki görüntünün toplam sınıflandırma doğruluğunu yaklaşık % 24 oranında arttırdığı tespit edilmiştir. Önerilen yöntem ile yüksek sınıflandırma doğruluğuna ihtiyaç duyulan uygulamalar için altlık üretilebileceği sonucuna varılmıştır.

References

  • Abd El-Samie, F. E., Hadhoud, M. M., & El-Khamy, S. E. (2012). Image Super-Resolution and Applications. CRC press.
  • Deb, K., Pratap, A., Agarwal, S., & Meyarivan, T. A. M. T. (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation, 6(2), 182-197. doi: 10.1109/4235.996017.
  • Garzelli, A., & Nencini, F. (2006a). Fusion of Panchromatic and Multispectral Images by Genetic Algorithms. IEEE International Conference on Geoscience and Remote Sensing Symposium (IGARSS 2006) (3810-3813). Denver, CO, USA. doi: 10.1109/IGARSS.2006.976.
  • Garzelli, A., & Nencini, F. (2006b). PAN-sharpening of Very High Resolution Multispectral Images Using Genetic Algorithms. International Journal of Remote Sensing, 27(15), 3273-3292. doi: 10.1080/01431160600554991.
  • Ghassemian, H. (2016). A review of remote sensing image fusion methods. Information Fusion, 32, 75-89. doi: 10.1016/j.inffus.2016.03.003.
  • Gogineni, R., & Chaturvedi, A. (2018). Sparsity inspired pan-sharpening technique using multi-scale learned dictionary. ISPRS Journal of Photogrammetry and Remote Sensing, 146, 360-372. doi: 10.1016/j.isprsjprs.2018.10.009.
  • Gonzalez, R. C., & Woods, R.E. 2007. Digital Image Processing. 3rd Edition, Pearson.
  • Laben, C. A., & Brower, B. V. (2000). U.S. Patent No. 6,011,875. Washington, DC: U.S. Patent and Trademark Office.
  • Serifoglu Yilmaz, C., Yilmaz, V., & Güngör, O. (2020). On the use of the SOS metaheuristic algorithm in hybrid image fusion methods to achieve optimum spectral fidelity. International Journal of Remote Sensing, 41(10), 3993-4021. doi: 10.1080/01431161.2019.1711244.
  • Serifoglu Yilmaz, C., Yilmaz, V., Gungor, O., & Shan, J. (2019). Metaheuristic pansharpening based on symbiotic organisms search optimization. ISPRS Journal of Photogrammetry and Remote Sensing, 158, 167-187. doi: 10.1016/j.isprsjprs.2019.10.014.
  • Tso, B., & Mather, P. 2009. Classification Methods for Remotely Sensed Data. 2nd Edition, CRC Press.
  • Wald, L. 2000. Quality of high resolution synthesized images: Is there a simple criterion? 3rd Conference: Fusion of Earth Data: Merging Point Measurements, Raster Maps and Remotely Sensed Images (99-103). Sophia Antipolis, France.
  • Yılmaz, V. (2020a). Metasezgisel Guguk Kuşu Arama Algoritması ile Görüntü Kaynaştırma. Türk Uzaktan Algılama ve CBS Dergisi, 1(1), 1-12.
  • Yilmaz, V. (2020b). A Non‐Dominated Sorting Genetic Algorithm‐II‐based approach to optimize the spectral and spatial quality of component substitution‐based pansharpened images. Concurrency and Computation: Practice and Experience, e6030.
  • Yilmaz, V., & Gungor, O. (2016). Determining the optimum image fusion method for better interpretation of the surface of the Earth. Norsk Geografisk Tidsskrift-Norwegian Journal of Geography, 70(2), 69-81. doi: 10.1080/00291951.2015.1126761.
  • Yilmaz, V., Serifoglu Yilmaz, C., Güngör, O., & Shan, J. (2020). A genetic algorithm solution to the gram-schmidt image fusion. International Journal of Remote Sensing, 41(4), 1458-1485. doi: 10.1080/01431161.2019.1667553.
  • Zhou, J., Civco, D. L., & Silander, J. A. (1998). A wavelet transform method to merge Landsat TM and SPOT panchromatic data. International Journal of Remote Sensing, 19(4), 743-757. doi: 10.1080/014311698215973.
There are 17 citations in total.

Details

Primary Language Turkish
Subjects Photogrammetry and Remote Sensing
Journal Section Research Articles
Authors

Volkan Yılmaz 0000-0003-0685-8369

Publication Date March 13, 2021
Submission Date December 10, 2020
Acceptance Date January 11, 2021
Published in Issue Year 2021

Cite

APA Yılmaz, V. (2021). Çok Amaçlı Metasezgisel Optimizasyon Algoritmaları ile Hibrit Pan-Keskinleştirme Yöntemlerinden Üretilen Görüntülerin Sınıflandırma Performanslarının Arttırılmasına Yönelik Bir Araştırma. Türk Uzaktan Algılama Ve CBS Dergisi, 2(1), 1-10. https://doi.org/10.48123/rsgis.838767

Creative Commons License
Turkish Journal of Remote Sensing and GIS (Türk Uzaktan Algılama ve CBS Dergisi), Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License ile lisanlanmıştır.