Transfer Öğrenme Teknikleri Kullanarak Nohut Çeşidi Sınıflandırma

Yıl 2024, Cilt: 14 Sayı: 1, 48 - 58, 29.04.2024

İbrahim Kılıç , Nesibe Yalçın

Öz

Tohum saflığı, tarım üretiminde verimi artırmak ve ürün kalite standartlarını karşılamak için oldukça önemlidir. Bu durum, tohum üreticilerinden dağıtıcılarına tarım endüstrisinin, tohum saflığına daha fazla önem vermesini gerektirmektedir. Bu da tohum çeşidi sınıflandırma ve ayırma yöntemlerine ihtiyacı artırmıştır. Çalışma kapsamında, dünyada en çok üretilen yemeklik baklagillerden biri olan nohudun çeşit sınıflandırması problemi ele alınmıştır. Sınıflandırma için 14 adet ön eğitimli derin öğrenme modeli kullanılmış ve model performansları karşılaştırılarak ilgili problem için en başarılı model(ler) tespit edilmeye çalışılmıştır. Başarımı en yüksek modeller VGG16 ve VGG19, sırasıyla %96.7 ve %97 test doğruluklarına sahiptir ve daha verimli, kaliteli ve sürdürülebilir tohum üretiminin sağlanması için önemli bir araç olabilirler.

Anahtar Kelimeler

Derin öğrenme, evrişimli sinir ağı, nohut sınıflandırma, transfer öğrenme

Teşekkür

T.C. Tarım ve Orman Bakanlığı Geçit Kuşağı Tarımsal Araştırmalar Enstitüsü Müdürlüğüne, Doğu Akdeniz Geçit Kuşağı Tarımsal Araştırmalar Enstitüsü Müdürlüğüne, Kahramanmaraş İl Tarım ve Orman Müdürlüğüne ve Kayseri Develi İlçe Tarım ve Orman Müdürlüğüne, nohut tohumlarının temini için teşekkür ederiz.

Chickpea Variety Classification Using Transfer Learning Techniques

Öz

Seed purity is important for improving the efficiency of agricultural production and meeting product quality standards. This requires the agricultural seed industry, from producers to distributors/sellers, to focus more on seed purity. Therefore, the need for seed variety identification and classification methods has increased. The seed variety classification of chickpeas, one of the most produced edible legumes in the world, is examined in this study. 14 pre-trained deep learning models have been used for classification and their performances have been compared to determine the most successful model(s) for the relevant problem. The most successful models, VGG16 and VGG19, have test accuracies of 96.7% and 97%, respectively. Thus, they can be important tools for ensuring more efficient, high-quality, and sustainable seed production.

Anahtar Kelimeler

Deep learning, convolutional neural network, chickpea classification, transfer learning

Kaynakça

• Abuhayi, BM., Bezabih, YA. 2023. Chickpea disease classification using hybrid method. Smart Agricultural Technology, 6:100371. Doi: 10.1016/j.atech.2023.100371
• Aktaş, H. 2022. Antep fıstığının derin öğrenme ile dış kabuk rengine göre sınıflandırılması. NÖHÜ Müh. Bilim. Derg., 11(3):461-466.
• Altan, G. 2019. DeepGraphNet: grafiklerin sınıflandırılmasında derin öğrenme modelleri. EJOSAT, 319-327. Doi: 10.31590/ejosat.638256
• Ayele, NA., Tamiru, HK. 2020. Developing classification model for chickpea types using machine learning algorithms. International Journal of Innovative Technology and Exploring Engineering (IJITEE), 10(1): 5-11. Doi: 10.35940/ijitee.A8057.1110120
• Başol, Y., Toklu, S. 2021. A deep learning-based seed classification with mobile application. Turk. J. Math. Comput. Sci., 13(1): 192-203. Doi: 10.47000/tjmcs.897631
• Chollet, F. 2017. Xception: deep learning with depthwise separable convolutions. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 1800-1807. Doi: 10.1109/CVPR.2017.195
• Çakmak, YS., Boyacı, İH. 2011. Quality evaluation of chickpeas using an artificial neural network integrated computer vision system. Int. J. Food Sci. Technol., 46(1): 194-200.
• Çetiner, H. 2022. Classification of plant species with transfer learning-based methods. EasyChair Preprint no. 8447.
• Géron A. 2019. Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition. Scikit-Learn, Keras ve TensorFlow ile Uygulamalı Makine Öğrenmesi. 1. Baskı. Çevirenler: Aksoy B., Özgür K., Buzdağı Yayınevi, İstanbul.
• Golcuk, A., Yasar, A., Saritas, MM., Erharman, A. 2023. Classification of cicer arietinum varieties using MobileNetV2 and LSTM. Eur. Food Res. Technol., 249:1343-1350. Doi: 10.1007/s00217-023-04217-w
• He, K., Zhang, X., Ren, S., Sun, J. 2016. Deep residual learning for image recognition. IEEE Conference on Computer Vision and Pattern Recognition, 770-778.
• Howard, AG., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., Adam, H. 2017. MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv Prepr., arXiv:1704.04861
• Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, KQ. 2017. Densely connected convolutional networks. IEEE Conference on Computer Vision and Pattern Recognition, 4700-4708.
• Karadağ, K., Keskinbıçak, F. 2023. Estimation chickpea species and productivity per decare with synthetic data generation methods, C. R. Acad. Bulg. Sci., 76(1): 146-155.
• Kaya, A., Çelik, A., Özkaya, U., Yigit, E. 2022. Derin öğrenme ile tahıl ayıklama, 1st International Conference on Engineering and Applied Natural Sciences (ICAENS 2022), Konya, Türkiye, 2501-2508.
• Kılıç, İ. 2024. TRCS_5_SET. https://github.com/ibrahimkilic/TRCS_5_SET
• Kılıç, İ., Yalçın, N. 2023. Evrişimsel sinir ağları tabanlı nohut çeşitleri sınıflandırması. Innovations in Intelligent Systems and Applications Conference (ASYU 2023), Sivas, Türkiye. IEEE. Doi: 10.1109/ASYU58738.2023.10296680
• LeCun, Y., Bengio, Y., Hinton, G. 2015. Deep learning. Nature, 521: 436-444. Doi: 10.1038/nature14539
• Narin, A., Kaya, C. Pamuk, Z. 2021. Automatic detection of coronavirus disease (COVID-19) using X-ray images and deep convolutional neural networks. Pattern Anal. Applic., 24: 1207-1220. Doi: 10.1007/s10044-021-00984-y
• Saha, D., Manickavasagan, A. 2022. Chickpea varietal classification using deep convolutional neural networks with transfer learning. J. Food Process Eng., 45(3): e13975. Doi: 10.1111/jfpe.13975
• Saha, D., Mangukia, MP., Manickavasagan, A. 2023. Real-time deployment of MobileNetV3 model in edge computing devices using RGB color images for varietal classification of chickpea. Appl. Sci., 13(3): 7804. Doi: 10.3390/app13137804
• Selvaraju, RR., Cogswell, M., Das, A., Vedantam, R., Parikh, D., Batra, D. 2017. Grad-CAM: visual explanations from deep networks via gradient-based localization, IEEE International Conference on Computer Vision, 618-626.
• Singh, D., Taspinar, YS., Kursun, R., Cinar, I., Koklu, M., Ozkan, İA., Lee, HN. 2022. Classification and analysis of pistachio species with pre-trained deep learning models. Electronics, 11: 981. Doi: 10.3390/electronics11070981
• Simonyan, K. Zisserman, A. 2015. Very deep convolutional networks for large-scale image recognition. International Conference on Learning Representations, 1-14.
• Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R. 2014. Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res., 15: 1929-1958.
• Szegedy, C., Ioffe, S., Vanhoucke, V., Alemi, AA. 2017. Inception-v4, inception-ResNet and the impact of residual connections on learning. AAAI Conference On Artificial Intelligence, San Francisco, California, USA, 4278-4284.
• Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A. 2015. Going deeper with convolutions. IEEE Conference on Computer Vision and Pattern Recognition, 1-9.
• Taheri-Garavand, A., Nasiri, A., Fanourakis, D., Fatahi, S., Omid, M., Nikoloudakis, N. 2021. Automated in situ seed variety identification via deep learning: a case study in Chickpea. Plants, 10(7): 1406. Doi: 10.3390/plants10071406
• Tan, M., Le, QV. 2019. EfficientNet: rethinking model scaling for convolutional neural networks. International Conference on Machine Learning, 6105-6114.
• Tuğrul, B., Sivari, E., Akca, S., Eryigit, R. 2022. Classification of dianthus seed species with deep transfer learning. SSRN. Doi: 10.2139/ssrn.4173707
• Xu, P., Yang, R., Zeng, T., Zhang, J., Zhang, Y., Tan, Q. 2021. Varietal classification of maize seeds using computer vision and machine learning techniques. J. Food Process Eng., 44(11): e13846. Doi: 10.1111/jfpe.13846
• Yaşar, A. 2023. Benchmarking analysis of CNN models for bread wheat varieties. Eur. Food Res. Technol., 249: 749-758. Doi: 10.1007/s00217-022-04172-y
• Zhang, A., Lipton, ZC., Li, M., Smola, AJ. 2023. Dive into deep learning. Cambridge University Press.
• Zhang, X., Zhou, X., Lin, M., Sun, J. 2018. ShuffleNet: an extremely efficient convolutional neural network for mobile devices. IEEE Conference on Computer Vision and Pattern Recognition, 6848-6856.
• Zoph, B., Le, QV. 2017. Neural architecture search with reinforcement learning. International Conference on Learning Representations.
• Zoph, B., Vasudevan, V., Shlens, J., Le, QV. 2018. Learning transferable architectures for scalable image recognition. IEEE Conference on Computer Vision and Pattern Recognition, 8697-8710.