UC Merced Land-Use Veri Kümesi Kullanılarak CNN Tabanlı Hibrit Sistemlerin Performans Karşılaştırması

Abstract

Uzaktan algılama, yeryüzü ile ilgili verilerin özel sensörler aracılığıyla toplanması ve incelenmesi teknolojisidir. Elde edilen veriler birçok uygulama alanında kullanılmaktadır. Uzaktan algılama görüntülerinin sınıflandırma başarısı, kullanılacak bilgilerin doğruluğu ve güvenilirliği ile yakından ilgilidir. Bu nedenle özellikle son yıllarda yapılan çalışmalarda birçok alanda popüler hale gelen Konvolüsyonel Sinir Ağlarının (CNN) kullanıldığı ve yüksek başarılar elde edildiği görülmektedir. Ancak bu bilgilerin hızlı bir şekilde elde edilmesi de önemli bir ihtiyaçtır. Dolayısıyla bu çalışmada CNN kadar başarılı ve CNN'den daha kısa sürede sonuç alınması amaçlanmaktadır. CNN ile özniteliklerin çıkarıldığı ve daha sonra makine öğrenmesi algoritmaları ile sınıflandırmanın yapıldığı hibrit sistemler test edilmiştir. İki farklı CNN mimarisi (ResNet18, GoogLeNet) ve dört farklı sınıflandırıcının (Support Vector Machine, K Nearest Neighbor, Decision Tree, Discriminant Analysis) ikili kombinasyonlarının başarıları çeşitli metriklerle karşılaştırılmıştır. GoogLeNet & SVM (%93,33) en yüksek doğruluk oranına sahip yöntem olurken, ResNet18 & DT (%50,95) en düşük doğruluk oranına sahip yöntemdir.

Keywords

• Uzaktan Algılama
• Konvolüsyonel Sinir Ağı
• Makine Öğrenmesi
• Hibrit Sistemler

Performance Comparison of CNN Based Hybrid Systems Using UC Merced Land-Use Dataset

Abstract

Remote sensing is the technology of collecting and examining data about the earth with special sensors. The data obtained are used in many application areas. The classification success of remote sensing images is closely related to the accuracy and reliability of the information to be used. For this reason, especially in recent studies, it is seen that Convolutional Neural Network (CNN), which has become popular in many fields, is used and high successes have been achieved. However, it is also an important need to obtain this information quickly. Therefore, in this study, it is aimed to get results as successful as CNN and in a shorter time than CNN. Hybrid systems in which features are extracted with CNN and then classification is performed with machine learning algorithms have been tested. The successes of binary combinations of two different CNN architectures (ResNet18, GoogLeNet) and four different classifiers (Support Vector Machine, K Nearest Neighbor, Decision Tree, Discriminant Analysis) have been compared with various metrics. GoogLeNet & Support Vector Machine (93.33%) is the method with the highest accuracy rate, while ResNet18 & Decision Tree (50.95%) is the method with the lowest accuracy rate.

Keywords

• Remote Sensing
• Convolutional Neural Network
• Machine Learning
• Hybrid Systems
• Uzaktan Algılama
• Konvolüsyonel Sinir Ağı
• Makine Öğrenmesi
• Hibrit Sistemler

References

1. [1] Yener, H., Koç, A., Çoban, H.O. 2006. Assessment Methods for Classification Accuracy in Remote Sensing Data, Journal of the Faculty of Forestry Istanbul University, 56(2), 71-88.
2. [2] Çölkesen, İ., Kavzoğlu, T. 2013. Arazi Örtüsü Haritalarinin Üretilmesinde Yeryüzü Özelliklerinin Siniflandirma Doğruluğuna Etkilerinin İncelenmesi: Trabzon örneği, Türkiye Ulusal Fotogrametri ve Uzaktan Algılama Birliği VII. Teknik Sempozyumu.
3. [3] Döş, M.E., Uysal, M. 2019. Classification of Remote Sensing Data with Deep Learning Algorithms, Turkish Journal of Remote Sensing, 1(1), 28-34.
4. [4] Sabanci, K., Ünlerşen, M.F. Kayabaşi, A. 2016. Machine Learning Methods for Land Cover Classification from Multispectral Images, IMCOFE, 472-478.
5. [5] Tavus, B., Karataş, K., Türker, M. 2018. Object-Based Crop Pattern Detection from IKONOS Satellite Images in Agricultural Areas, Pamukkale University Journal of Engineering Sciences, 25(5), 603-614.
6. [6] Özyurt, F. 2019. Classification of Remote Sensing Images Based on Convolutional Neural Networks and Neighborhood Component Analysis Features, Afyon Kocatepe University Journal of Science and Engineering, 19(3), 669-675.
7. [7] Bilgilioğlu, B.B., Çömert, R., Yiğit, O., Bedir F. 2019. Extraction of Tea Gardens by Object-Based Classifıcation Approach From High Spatial Resolution Satellite Images, Turkish Journal of Remote Sensing, 1(1), 21-27.
8. [8] Çan, T., Tekin, S., Traore, M., & Kumsar, H. 2020. Land Use Change Detection in Denizli City Center Using spectral angle mapper method and evaluations in terms of some earth science data, Pamukkale University Journal of Engineering Sciences, 26(8), 1360-1364.

9. [9] Apaydin, C., Abdikan, S., 2021. Determination of Hazelnut Gardens by Pixel Based Classification Methods Using Sentinel-2 Data, Geomatik, 6(2), 107-114.
10. [10] Akram, T., Laurent, B., Naqvi, S.R., Alex, M.M., Muhammad, N. 2018. A Deep Heterogeneous Feature Fusion Approach for Automatic Land-use Classification, Information Sciences, 467, 199-218.
11. [11] Yuan, B., Li, S., Li, N. 2018. Multiscale Deep Features Learning for Land-Use Scene Recognition, Journal of Applied Remote Sensing, 12(1).
12. [12] Shafaey, M.A., Salem, M.A.M., Ebeid, H.M., Al-Berry, M.N., Tolba M.F. 2018. Comparison of CNNs for Remote Sensing Scene Classification, 13th International Conference on Computer Engineering and Systems, 27-32.
13. [13] Iorga, C., Neagoe, V.E. 2019. A Deep CNN Approach with Transfer Learning for Image Recognition, 11th International Conference on Electronics, Computers and Artificial Intelligence, 1-6.
14. [14] Xu, L., Chen, Y., Pan, J., Gao, A. 2020. Multi-Structure Joint Decision-Making Approach for Land Use Classification of High-Resolution Remote Sensing Images Based on CNNs, IEEE Access, 8, 42848-42863.
15. [15] Zhang, Z., Cui, X., Zheng, Q., Cao, J. 2021. Land Use Classification of Remote Sensing Images Based on Convolution Neural Network, Arabian Journal of Geosciences, 14(4), 1-6.
16. [16] Yang, Y., Newsam, S. 2010. Bag-of-Visual-Words and Spatial Extensions for Land-Use Classification, Proceedings of the 18th SIGSPATIAL international conference on advances in geographic information systems, 270-279.
17. [17] Karahan, T., Nabiyev, V. 2021. Plant Identification with Convolutional Neural Networks and Transfer Learning, Pamukkale University Journal of Engineering Sciences, 27(5), 638-645.
18. [18] Özkan, K., Seke, E., Işık, Ş. 2021. Wheat Kernels Classification Using Visible-Near Infrared Camera Based on Deep Learning, Pamukkale University Journal of Engineering Sciences, 27(5), 618-626.
19. [19] Cevik, F., Kilimci, Z.H. 2021. The Evaluation of Parkinson's Disease with Sentiment Analysis using Deep Learning Methods and Word Embedding Models, Pamukkale University Journal of Engineering Sciences, 27(2), 151-161.
20. [20] Çelik, G., Talu, M.F. 2021. Generating the Image Viewed from EEG Signals, Pamukkale University Journal of Engineering Sciences, 27(2), 129-138.
21. [21] LeCun, Y., Bottou, L., Bengio, Y., Haffner, P. 1998. Gradient-Based Learning Applied to Document Recognition, Proceedings of the IEEE, 86(11), 2278-2324.
22. [22] Krizhevsky, A., Sutskever, I., Hinton, G.E. 2012. Imagenet Classification with Deep Convolutional Neural Networks, Advances in neural information processing systems, 25, 1097-1105.
23. [23] Zeiler, M.D., Fergus, R. 2014. Visualizing and Understanding Convolutional Networks, In European conference on computer vision, 818-833.
24. [24] Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Rabinovich, A. 2015. Going Deeper with Convolutions, In Proceedings of the IEEE conference on computer vision and pattern recognition, 1-9.
25. [25] Simonyan, K., Zisserman, A. 2014. Very Deep Convolutional Networks for Large-Scale Image Recognition, arXiv preprint arXiv:1409.1556.
26. [26] He, K, Zhan,g X, Ren, S, Sun, J. 2016. Deep Residual Learning for Image Recognition, Proceedings of the IEEE conference on computer vision and pattern recognition, 770-778.
27. [27] Cortes, C., & Vapnik, V. 1995. Support-vector networks. Machine learning, 20, 273-297.
28. [28] Wang, M, Zhang, X, Niu, X, Wang, F, Zhang, X. 2019. Scene classification of high-resolution remotely sensed image based on ResNet. Journal of Geovisualization and Spatial Analysis, 3, 1-9.
29. [29] Fix E., & Hodges, J. 1951. Discriminatory Analysis: Nonparametric Discrimination: Consistency Properties, 4.
30. [30] Cover, T., & Hart, P. 1967. Nearest neighbor pattern classification. IEEE transactions on information theory, 13(1), 21-27.
31. [31] Bouteldja, S, Kourgli, A. 2020. A comparative analysis of SVM, K-NN, and decision trees for high resolution satellite image scene classification. In Twelfth International Conference on Machine Vision (ICMV 2019), 11433, 410-416.
32. [32] Quinlan, J. R. 1987. Simplifying decision trees. International journal of man-machine studies, 27(3), 221-234.
33. [33] Fisher, R. A. 1936. The use of multiple measurements in taxonomic problems. Annals of eugenics, 7(2), 179-188.
34. [34] Huang, R, Taubenböck, H, Mou, L, Zhu, X X 2018. Classification of settlement types from Tweets using LDA and LSTM. In IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, 6408-6411.
35. [35] Negrel, R., Picard, D., Gosselin, P.H. 2014. Evaluation of Second-Order Visual Features for Land-Use Classification, 12th International Workshop on Content-Based Multimedia Indexin, 1-5.
36. [36] Alias, B., Karthika, R., Parameswaran, L. 2018. Classification of High Resolution Remote Sensing Images Using Deep Learning Techniques, International Conference on Advances in Computing, Communications and Informatics, 1196-1202.
37. [37] Helber, P., Bischke, B., Dengel, A., Borth, D. 2019. Eurosat: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 12(7), 2217-2226.