Ön-Eğitimli CNN Mimarilerinin Füzyonu ile Mısır Yaprağı Hastalıklarının Sınıflandırılması

Abstract

Mısır, dünyanın birçok yerinde temel gıda olarak kullanılmaktadır. Mısır, yağ ve yüksek fruktozlu mısır şurubu yapmak için kullanılabilecek iyi bir nişasta kaynağıdır. Mısır ve mısır unu çok hayati ürünler olduğundan, Yanık, Pas ve Gri Yaprak Lekesi gibi bazı hastalıkların erken teşhisiyle bitkilerin iyileştirilmesini sağlanarak ürün kalitesinin düşmesi önlenebilecektir. Bu sayede hem ürünün kalitesi hem de elde edilen ürün miktarı arttırılabilmektedir. Bu çalışmada, Evrişimli Sinir Ağı (CNN) tabanlı VGG-19, DenseNet-201 ve NasNet-Large modelleri kullanılarak mısır yaprağı görüntülerinden öznitelik çıkarımı yapılmıştır. Veri setindeki dengesizliğin giderilmesi için Sentetik Azınlık Yüksek Örnekleme Tekniği (SMOTE) yöntemi ile balans ayarı yapılmıştır. Öznitelik sayısını düşürmek için boyut indirgeme yöntemlerinden Temel Bileşen Analizi (PCA) kullanılmıştır. Mısır yaprağındaki hastalıkları sınıflandırmak amacıyla Destek-Vektör Makinaları (SVMs) algoritması kullanılmıştır. Algoritmanın performansını artırmaya yönelik GridSearchCV yaklaşımı ile mısır yapraklarındaki hastalıkları tanımlamak için Kernel function ve Box constrain hiperparametreleri optimize edilmiştir. Elde edilen deneysel sonuçlar genel erişime açık Kaggle mısır veya mısır yaprağı hastalığı veri kümesi üzerinde test edilmiştir. Elde edilen deneysel sonuçlarda sadece CNN ile özellik çıkarımı yapılan görüntülerin LibSVM ile sınıflandırılmasında 4 sınıf için sırasıyla %94,5, %94,4, %94,3, ve %96,2 doğruluk oranlarına ve %94,3 ağırlıklı ortalamaya ulaşılmıştır. Önerilen yöntem kullanılarak LibSVM ile 4 sınıf için sırasıyla %96,7, %96,7, %96,7 ve %97,8 doğruluk oranlarına ve %96,7 ağırlık ortalamaya ulaşılmıştır. Böylece önerilen yöntemle elde edilen sınıflandırma doğruluğunda optimizasyon yapılmadan elde edilen sınıflandırma doğruluğuna göre birinci sınıf için %2,2, ikinci sınıf için %2,3, üçüncü sınıf için %2,4 ve dördüncü sınıf için %1,6, bununla birlikte ağırlıklı ortalamada %2,4 oranında iyileşme sağlandığı görülmüştür.

Keywords

• CNN
• LibSVM
• Mısır yaprağı
• Optimizasyon
• PCA
• Sınıflandırma

Classification of Maize Leaf Diseases by Fusion of Pre-Trained CNN Architectures

Abstract

Maize is used as a staple food in many parts of the world. Maize is a good source of starch that can be used to make oil and high fructose corn syrup. Since maize and maize flour are very vital products, early diagnosis of some diseases such as Blight, Rust, and Gray Leaf Spot can prevent the deterioration of product quality by improving the plants. In this way, both the quality of the product and the amount of product obtained can be increased. In this study, feature extraction was performed from corn leaf images using Convolutional Neural Network (CNN) based VGG-19, DenseNet-201, and NasNet-Large models. In order to eliminate the imbalance in the data set, the balance was adjusted with the Synthetic Minority Over-Sampling Technique (SMOTE) method. Principal Component Analysis (PCA), one of the dimension reduction methods, was used to reduce the number of features. Support-Vector Machines (SVMs) algorithm was used to classify diseases in maize leaves. With the GridSearchCV approach to improving the performance of the algorithm, the Kernel function and Box constrain hyperparameters have been optimized to identify diseases in corn leaves. The experimental results obtained were tested on the publicly accessible Kaggle corn or maize leaf disease dataset. In the experimental results obtained, 94.5%, 94.4%, 94.3%, and 96.2% accuracy rates and a weighted average of 94.3% were achieved for the 4 classes, respectively, in the classification of the images with only CNN and feature extraction with LibSVM. Using the proposed method, 96.7%, 96.7%, 96.7%, and 97.8% accuracy rates and 96.7% weight average were achieved for 4 classes with LibSVM, respectively. Thus, according to the classification accuracy obtained without optimization in the classification accuracy obtained with the proposed method, it was observed that there was an improvement of 2.2% for the first class, 2.3% for the second class, 2.4% for the third class, and 1.6% for the fourth class, and 2.4% in the weighted average.

Keywords

• CNN
• LibSVM
• Mısır yaprağı
• Optimizasyon
• PCA
• Sınıflandırma

References

1. Akpınar, H. (2014). Data: Veri Madenciliği Veri Analizi, 1. baskı. Papatya Yayıncılık Eğitim, İstanbul.
2. Al-Amin, M., Karim, D. Z., & Bushra, T. A. (2019, December). Prediction of rice disease from leaves using deep convolution neural network towards a digital agricultural system. In 2019 22nd International Conference on Computer and Information Technology (ICCIT) (pp. 1-5). IEEE.
3. Albawi, S., Mohammed, T. A., & Al-Zawi, S. (2017, August). Understanding of a convolutional neural network. In 2017 international conference on engineering and technology (ICET) (pp. 1-6). Ieee.
4. Ali, H., Maulud, A. S., Zabiri, H., Nawaz, M., Suleman, H., & Taqvi, S. A. A. (2022). Multiscale Principal Component Analysis-Signed Directed Graph Based Process Monitoring and Fault Diagnosis. ACS omega, 7(11), 9496-9512.
5. Arık, A. O. (2021, Jan 14). Medium: https://okanarik.medium.com/smote-synthetic-minority-oversampling-technique-c52d4fbec47e. Erişim: 04.10.2022
6. Atallah, R., & Al-Mousa, A. (2019, October). Heart disease detection using machine learning majority voting ensemble method. In 2019 2nd international conference on new trends in computing sciences (ictcs) (pp. 1-6). IEEE.
7. Bhange, M., & Hingoliwala, H. A. (2015). Smart farming: Pomegranate disease detection using image processing. Procedia computer science, 58, 280-288.
8. Campus, P. (2012). Inoculation methods and disease rating scales for maize diseases. (Revised). Directorate of Maize Research, New Delhi.

9. Chen, L., & Wang, L. Y. (2011). Research on application of probability neural network in maize leaf disease identification. J. Agricult. Mech. Res, 33(6), 145-148.
10. DeChant, C., Wiesner-Hanks, T., Chen, S., Stewart, E. L., Yosinski, J., Gore, M. A., ... & Lipson, H. (2017). Automated identification of northern leaf blight-infected maize plants from field imagery using deep learning. Phytopathology, 107(11), 1426-1432.
11. Dillard, H. R., & Seem, R. C. (1990). Incidence-severity relationships for common maize rust on sweet corn. Phytopathology, 80(9), 842-849.
12. Dixit, A., & Nema, S. (2018). Wheat leaf disease detection using machine learning method-a review. Int. J. Comput. Sci. Mob. Comput, 7(5), 124-129.
13. Fan, P. H. C. R. E., & Lin, C. J. (2005). Dimensionality Reduction via Sparse Support Vector Machines. Journal of Machine Learning Research, 6, 1889-1918.
14. Floridi, L. (2020). AI and its new winter: From myths to realities. Philosophy & Technology, 33(1), 1-3. Geekycodesco. (2022, Mart 13). By geekycodesco: https://geekycodes.in/what-is-the-vgg-19-neural-network/ Erişim: 28.09.2022
15. Ghose, S. (2022). Kaggle. Kaggle: https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset. Erişim: 17.09.2022
16. Hao, W., & Zhang, Z. (2019). Spatiotemporal distilled dense-connectivity network for video action recognition. Pattern Recognition, 92, 13-24.
17. Huang, G. B., Zhu, Q. Y., & Siew, C. K. (2004, July). Extreme learning machine: a new learning scheme of feedforward neural networks. In 2004 IEEE international joint conference on neural networks (IEEE Cat. No. 04CH37541) (Vol. 2, pp. 985-990). Ieee.
18. Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017). Densely connected convolutional networks Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2017. Honolulu, HI.[Google Scholar].
19. Kapon, O. (2022, February). Kaggle: https://www.kaggle.com/code/omreekapon/corn-and-maize-diseases-classification/notebook. Erişim: 04.10.2022
20. Kılıç, S. (2013). Klinik karar vermede ROC analizi. Journal of Mood Disorders, 3(3), 135-40.
21. Kırtok, Y. (1998). Mısır: üretimi ve kullanımı. Kocaoluk Yayınevi.
22. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). Imagenet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84-90.
23. Kusumo, B. S., Heryana, A., Mahendra, O., & Pardede, H. F. (2018, November). Machine learning-based for automatic detection of corn-plant diseases using image processing. In 2018 International conference on computer, control, informatics and its applications (IC3INA) (pp. 93-97). IEEE.
24. Leung, K. (2021, Jan 4). Towards Data Science: https://towardsdatascience.com/micro-macro-weighted-averages-of-f1-score-clearly-explained-b603420b292f. Erişim: 03.10.2022
25. Li, C., & Lanying, W. (2011). Research on Application of Probability Neural Network in Maize Leaf Disease Identification [J]. J Agric Mechan Res, 6.
26. Lu, Y., Yi, S., Zeng, N., Liu, Y., & Zhang, Y. (2017). Identification of rice diseases using deep convolutional neural networks. Neurocomputing, 267, 378-384.
27. Lumini, A., & Nanni, L. (2019). Deep learning and transfer learning features for plankton classification. Ecological informatics, 51, 33-43.
28. Metlek, S., & Kayaalp, K. (2020). Makine Öğrenmesinde, Teoriden Örnek MATLAB Uygulamalarına Kadar Destek Vektör Makineleri. İksad Yayınevi.
29. Miglani, V., & Bhatia, M. P. S. (2020, February). Skin lesion classification: A transfer learning approach using efficientnets. In International Conference on Advanced Machine Learning Technologies and Applications (pp. 315-324). Springer, Singapore.
30. Miller, S. A., Beed, F. D., & Harmon, C. L. (2009). Plant disease diagnostic capabilities and networks. Annual review of phytopathology, 47(1), 15-38.
31. Mosetti, R. (2016). Principal Component Analysis of quantum correlation. The European Physical Journal Plus, 131(12), 1-8.
32. Muratlar, E. R. (2021, Temmuz 9). Dengesiz Veri Setlerinde Modelleme. Veri Bilimi: https://www.veribilimiokulu.com/dengesiz-veri-setlerinde-modelleme/#:~:text=SMOTE(Synthetic%20Minority%20Over%2DSampling,yeni%20az%C4%B1nl%C4%B1k%20s%C4%B1n%C4%B1f%C4%B1%20%C3%B6rnekleri%20yaratmakt%C4%B1r. Erişim: 05.10.2022
33. Öğündür, G. (2019, Nov 9). Medium: https://medium.com/@gulcanogundur/do%C4%9Fruluk-accuracy-kesinlik-precision-duyarl%C4%B1l%C4%B1k-recall-ya-da-f1-score-300c925feb38. Erişim: 02.10.2022
34. Panigrahi, K. P., Sahoo, A. K., & Das, H. (2020, June). A cnn approach for corn leaves disease detection to support digital agricultural system. In 2020 4th International Conference on Trends in Electronics and Informatics (ICOEI)(48184) (pp. 678-683). IEEE.
35. Pujari, D., Yakkundimath, R., & Byadgi, A. S. (2016). SVM and ANN based classification of plant diseases using feature reduction technique. IJIMAI, 3(7), 6-14.
36. Qi, Z., Jiang, Z., Yang, C., Liu, L., & Rao, Y. (2016). Identification of maize leaf diseases based on image technology. Journal of Anhui Agricultural University, 43(2), 325-330.
37. Rage, S. (2022). Kaggle: https://www.kaggle.com/code/sailikhitarage/maize-mutant-classification-using-vgg16/notebook. Erişim: 04.10.2022
38. Sibiya, M., & Sumbwanyambe, M. (2019). A computational procedure for the recognition and classification of maize leaf diseases out of healthy leaves using convolutional neural networks. AgriEngineering, 1(1), 119-131.
39. Song, K., Sun, X. Y., & Ji, J. W. (2007). Corn leaf disease recognition based on support vector machine method. Transactions of the CSAE, 23(1), 155-157.
40. Şahin, S. (2001). Türkiyede Mısır Ekim Alanlarının Dağılışı Ve Mısır Üretimi. Gazi Üniversitesi Gazi Eğitim Fakültesi Dergisi, 21(1).
41. Wang, N., Wang, K., Xie, R., Lai, J., Ming, B., & Li, S. (2009). Maize leaf disease identification based on fisher discrimination analysis. Scientia Agricultura Sinica, 42(11), 3836-3842.
42. Ward, J. M., Stromberg, E. L., Nowell, D. C., & Nutter Jr, F. W. (1999). Gray leaf spot: a disease of global importance in maize production. Plant disease, 83(10), 884-895.
43. Weizheng, S., Yachun, W., Zhanliang, C., & Hongda, W. (2008, December). Grading method of leaf spot disease based on image processing. In 2008 international conference on computer science and software engineering (Vol. 6, pp. 491-494). IEEE.
44. Whxna. (2022, August). Kaggle: https://www.kaggle.com/code/jiaowoguanren/corn-or-maize-leaf-dataset-tf-squeezenet/notebook. Erişim: 04.10.2022
45. Xu, L., Xu, X., Hu, M., Wang, R., Xie, C., & Chen, H. (2015). Corn leaf disease identification based on multiple classifiers fusion. Transactions of the Chinese Society of Agricultural Engineering, 31(14), 194-201.
46. Zhang, F. (2013). Recognition of corn leaf disease based on quantum neural network and combination characteristic parameter. Journal of Southern Agriculture, 44(8), 1286-1290.
47. Zhang, Z. Y., He, X. Y., Sun, X. H., Guo, L. M., Wang, J. H., & Wang, F. S. (2015). Image recognition of maize leaf disease based on GA-SVM. Chemical Engineering Transactions, 46, 199-204.
48. Zoph, B., Vasudevan, V., Shlens, J., & Le, Q. V. (2018). Learning transferable architectures for scalable image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 8697-8710).