Bitki Sınıflandırması için Transfer Learning Kullanılarak Topluluk Öğrenmesi Metodu Üzerine Bir Çalışma

Öz

Derin öğrenme, insana özgü problemlerin gelişmiş donanım gücüne sahip makineler yardımıyla çözüldüğü önemli bir disiplindir. Bu disiplinin sanayi, sağlık, savunma sanayi ve spor alanlarında yaygın olarak kullanıldığı görülmektedir. Ayrıca bahçecilik alanında derin öğrenmenin kullanılması önemli bir gerekliliktir. Derin öğrenmenin bahçeciliğe entegrasyonu ile ürün sınıflandırması yapmak, verimliliği ve üretimi artırmak için oldukça önemlidir. Bu çalışmada çeşitli bitki verilerini kullanarak sınıflandırma probleminin doğruluğunu artırmak için topluluk öğrenmesi yöntemi önerilmiştir. Bu yöntem için veri artırmadan bağımsız olarak toplam 24421 görüntü ve 15 ürün sınıfı içeren yeni bir veri seti oluşturulmuştur. Önerilen yöntem yardımıyla oluşturulan bu veri setini eğitmek için bir modelin çıktısının diğer modelin girdisi olduğu hiyerarşik bir yapı tasarlanmıştır. Önerilen yöntemin deneysel çalışmalarında toplam 7 adet önceden eğitilmiş model kullanılmıştır. Bu yöntem bir topluluk yapısında olduğu için yapıya önceden eğitilmiş modeller eklemek veya çıkarmak mümkündür. Deneysel çalışmalar yardımıyla önerilen yöntemin geleneksel CNN yöntemi ile karşılaştırılan performans analizi yapılmıştır. Bu analizler sonucunda önerilen yöntemin %3 daha başarılı çalıştığı görülmüştür.

Anahtar Kelimeler

• Transfer learning
• ensemble learning
• convolutional neural network
• image classification
• deep learning.

A Study of Ensemble Deep Learning Method Using Transfer Learning for Horticultural Data Classification

Öz

Deep learning is an important discipline in which human-specific problems are solved with the help of machines with advanced hardware power. It is seen this discipline is widely used in the fields of industry, health, defense industry, and sports. In addition, the use of deep learning in the field of horticulture is an important requirement. With the integration of deep learning into horticulture, to do product classification is very important for increasing productivity and production. In this study, a method using ensemble learning is proposed to improve the accuracy of the classification problem for horticultural data. For this method, a new dataset was created, containing a total of 24421 images and 15 crop classes, independent of data augmentation. In order to train this created data set with the help of the proposed method, a hierarchical structure has been designed in which the output of one model is the input of the other model. A total of 7 pre-trained models were used in the experimental studies of the proposed method. Since this method is in an ensemble structure, it is possible to add or remove pre-trained models from the structure. With the help of experimental studies, a performance analysis of the proposed method, which is compared with the traditional CNN method, has been made. As a result of these analyses, it has been observed that the proposed method works 3% more successfully.

Anahtar Kelimeler

• Transfer learning
• ensemble learning
• convolutional neural network
• image classification
• deep learning.

Kaynakça

1. A. Palaparthi, A. M. Ramiya, H. Ram and D. D. Mishra, 2023. Classification of Horticultural Crops in High Resolution Multispectral Imagery Using Deep Learning Approaches, International Conference on Machine Intelligence for GeoAnalytics and Remote Sensing (MIGARS), Hyderabad, India.
2. Abed, S. H., Al Waisy, A. S., Mohammed, H. J., & Al Fahdawi, S., (2021). A modern deep learning framework in robot vision for automated bean leaves diseases detection, International Journal of Intelligent Robotics and Applications, 5, 235-251.
3. Ahmad, F., Farooq, A., & Ghani, M. U., (2021). Deep Ensemble Model for Classification of Novel Coronavirus in Chest X-Ray Images, Computational Intelligence and Neuroscience.
4. Altaf, F., Islam, S. M. S., & Janjua, N. K., (2021). A novel augmented deep transfer learning for classification of COVID-19 and other thoracic diseases from X-rays, Neural Computing and Applications.
5. Babu, S. A., &Annavarapu, C. S. R., (2021). Deep learning-based improved snapshot ensemble technique, The International Journal of Applied Intelligence, 51, 3104-3120.
6. Biswas, D., Su, H., Wang, C., Stevanovic, A., & Wang, W., (2018) An Automatic Traffic Density Estimation Using Single Shot Detection (SSD) and MobileNet-SSD, Physics and Chemistry of the Earth.
7. Bosilj, P., Aptoula, E., Duckett, T., &Cielniak, G., (2019). Transfer learning between crop types for semantic segmentation of crops versus weeds in precision agriculture, Journal of Field Robotics, 1-13.
8. D. Adams, (1979). The Hitchhiker's Guide to the Galaxy, London: Alfa.

9. Dawei, W., Limiao, D., Jiangong, N., Jiyue, G., Hongfei, Z., &Zhongzhi, H., (2019). Recognition pest by image-based transfer learning, Journal of the Science of Food and Agriculture, 99, 4524-4531.
10. Ganaiea, M., Hub, M., Tanveera, M., &Suganthanb, P., (2021). Ensemble deep learning: A review, Preprint submitted to Elsevier.
11. Garcia, B. E., Mylonas, N., Athanasakos, L., & Fountas, S., (2020). Towards weeds identification assistance through transfer learning, Computers and Electronics in Agriculture, 171.
12. Garcia, B. E., Mylonas, N., Athanasakos, L., Vali, E., & Fountas, S., (2021). Combining generative adversarial networks and agricultural transfer learning for weeds identification, ScienceDirect, 79-89.
13. Jahanbakhshi, A., Momeny, M. M., Mahmoudi, M., & Zhang, Y. D., (2020). Classification of sour lemons based on apparent defects using stochastic pooling mechanism in deep convolutional neural networks, Scientia Horticulture, 263.
14. Kaggle, Kaggle Inc, [Online]. Available: https://www.kaggle.com/. (Accessed: 04. Jul. 2022).
15. Kang, J., & Gwak, J., (2021). Ensemble of multitask deep convolutional neural networks using transfer learning for fruit freshness classification, Multimedia Tools and Applications.
16. LeCun, Y., Bengio, Y., & Hinton, G., 2015. Deep Learning, Nature, 521, 436-444.
17. Li, Y., Huang, H., Xie, Q., Yao, L., & Chen, Q., (2018). Research on a Surface Defect Detection Algorithm Based on MobileNet-SSD, Applied Sciences, 8(9), 1677-1694.
18. Linguo, L., Li, S., & Su, J., (2021). A Multi-Category Brain Tumor Classification Method Bases on Improved ResNet50, Computers, Materials & Continua, 2(69), 2355-2366.
19. Re, M., & Valentini, G., (2014). Ensemble methods: A review, Advances in Machine Learning and Data Mining for Astronomy, 563-594.
20. Salama, W. M., & Aly, M. H., (2021). Deep learning in mammography images, Alexandria Engineering Journal, 60, 4701-4709.
21. Shabbir, A., Ali, N., Ahmed, J., Zafar, B., Rasheed, A., Sajid, M., Ahmed, A., & Dar, S. H., (2021). Satellite and Scene Image Classification Based on Transfer Learning and Fine Tuning of ResNet50, Mathematical Problems in Engineering.
22. Tian, X., & Chen, C., (2019). Modulation Pattern Recognition Based on Resnet50 Neural Network, IEEE International Conference on Information Communication and Signal Processing, Beijing.
23. Vidal, P. L., Moura, J. d., Novo, J., & Orgeta, M., (2021). Multi-stage transfer learning for lung segmentation using portable X-ray, Expert Systems with Applications, 173.
24. Weiss, K., Khoshgoftaar, T. M., & Wang, D., (2016). A Survey of Transfer Learning, Journal of Big Data, 3, 9.
25. Xie, W., Wei, S., Zheng, Z., Jiang, Y., & Yang, D., (2021). Recognition of Defective Carrots Based on Deep Learning Deep Learning and Transfer Learning, Food and Bioprocess Technology, 14(7),1-14.
26. Yang, B., Xu, Y., 2021. Applications of deep-learning approaches in horticultural research: a review, Horticulture Research, p., 01 06 2021.
27. Yang, M., He, Y., Zhang, H., Li, D., Bouras, A., Yu, X., & Tang, Y., (2019). The Research on Detection of Crop Diseases Ranking Based on Transfer Learning, International Conference on Information Science and Control Engineering (ICISCE), Shanghai.
28. Zhao, W., Yamada, W., Li, T., Diagman, M., & Runge, T., (2021). Augmenting Crop Detection for Precision Agriculture with Deep Visual Transfer Learning A Case Study of Bale Detection, Remote Sensing, 13(23).