Enhancing agricultural classification models through data augmentation and advanced deep learning techniques
, &
Article Sidebar
Main Article Content
Abstract
In the field of agricultural data analysis, achieving high quality classification modeling remains a significant challenge due to the inherent variability and complexity of agricultural datasets. This study investigated cutting-edge approaches to enhance model performance through data augmentation techniques and the application of advanced deep learning models to artificially enlarge the training dataset, thereby improving model generalizability and robustness. Additionally, the study evaluated the efficacy of state-of-the-art models (i.e., ViT-Ti/16, CaiT-XXS-24, XCiT-T12, Resnet26, ConvNeXt-T) for agricultural data analysis. The experimental results revealed a marked improvement in terms of accuracy and F1-Score when applied data augmentation into the training session. This underscored the potential of these techniques to significantly advance the field of agricultural informatics. Briefly, the findings contributed to the development of more reliable and high performance models for agricultural practices.
Article Details
References
Bhuyan, P., & Singh, P. K. (2024). Evaluating deep CNNs and vision transformers for plant leaf disease classification. In Devismes, S., Mandal, P. S., Saradhi, V. V., Prasad, B., Molla, A. R., & Sharma, G. (Eds.), Proceedings of The 20th International Conference on Distributed Computing and Intelligent Technology ICDCIT 2024, Bhubaneswar, India, January 17-20, 2024 (293-306). Zug, Switzerland: Springer Cham. https://doi.org/10.1007/978-3-031-50583-6_20.
Cubuk, E. D., Zoph, B., Shlens, J., & Le, Q. V. (2020). Randaugment: Practical automated data augmentation with a reduced search space. In Boult, T., Medioni, G., & Zabih, R. (Eds.), Proceedings of The 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Seattle, WA, USA, June 14-19 (3008-3017). New Jersey, USA: Institute of Electrical and Electronics Engineers - IEEE. https://doi.org/10.1109/cvprw50498.2020.00359.
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., & Houlsby, N. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. arXiv 2010, 11929. https://doi.org/10.48550/arXiv.2010.11929.
El-Nouby, A., Touvron, H., Caron, M., Bojanowski, P., Douze, M., Joulin, A., Laptev, I., Neverova, N., Synnaeve, G., Verbeek, J., & Jegou, H. (2021). Xcit: Cross-covariance image transformers. arXiv 2106, 09681v2. https://doi.org/10.48550/arXiv.2106.09681.
He, M. K., Zhang, G. X., Ren, Q. S., & Sun, J. (2015). Deep residual learning for image recognition. In Tuytelaars, T., Li, F. F., & Bajcsy, R. (Eds.), Proceedings of The 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, June 27-30 (770-778). New Jersey, USA: Institute of Electrical and Electronics Engineers - IEEE. https://doi.org/10.1109/cvpr.2016.90.
Huang, G., Sun, Y., Liu, Z., Sedra, D., & Weinberger, K. (2016). Deep networks with stochastic depth. In Leibe, B., Sebe, N., Matas, J., & Welling, M. (Eds.), Proceedings of Computer Vision - ECCV 2016: 14th European Conference Part IV, Amsterdam, The Netherlands, October 11-14 (646-661). Zug, Switzerland: Springer Cham. https://doi.org/10.1007/978-3-319-46493-0_39.
Liu, Z., Mao, H., Wu, C. Y., Feichtenhofer, C., Darrell, T., & Xie, S. (2022). A convnet for the 2020s. In Chellappa, R., Matas, J., Quan, L., & Shah, M. (Eds.), Proceedings of The 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops New Orleans, Louisiana, June 19-24, 2022 (11976-11986). New Jersey, USA: Institute of Electrical and Electronics Engineers - IEEE. https://doi.org/10.1109/CVPR52688.2022.01167.
Phan, L. D., & Tran, T. S. (2022). Applying convolution neural networks for leaf image recognition with the vietnamese leaf image database. In Proceedings of The 4th International Conference on Sustainable Agriculture and Environment (81-94). Ho Chi Minh City, Vietnam: Nong Lam University.
Shorten, C., & Khoshgoftaar, T. M. (2019). A survey on image data augmentation for deep learning. Journal of Big Data 6(1), 1-48. https://doi.org/10.1186/s40537-019-0197-0.
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2015). Rethinking the inception architecture for computer vision. In Tuytelaars, T., Li, F. F., & Bajcsy, R. (Eds.), Proceedings of The 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, June 27-30 (2818-2826). New Jersey, USA: Institute of Electrical and Electronics Engineers - IEEE. https://doi.org/10.1109/cvpr.2016.308.
Touvron, H., Cord, M., Sablayrolles, A., Synnaeve, G., & Jégou, H. (2021). Going deeper with image transformers. In Berg, T., Clark, J., Matsushita, Y., & Taylor, C. J. (Eds.), Proceedings of The 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada, October 11-17 (32-42). New Jersey, USA: Institute of Electrical and Electronics Engineers - IEEE. https://doi.ieeecomputersociety.org/10.1109/ICCV48922.2021.00010.
Yun, S., Han, D., Oh, S. J., Chun, S., Choe, J., & Yoo, Y. (2019). CutMix: Regularization strategy to train strong classifiers with localizable features. In Lee, K. M., Forsyth, D., Pollefeys, M., & Tang, X. (Eds.), Proceedings of The 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, South Korea, October 27-November 2, (6022-6031). New Jersey, USA: Institute of Electrical and Electronics Engineers - IEEE. https://doi.ieeecomputersociety.org/10.1109/ICCV.2019.00612.
Zhang, H., Cisse, M., Dauphin, Y. N., & LopezPaz, D. (2017). Mixup: Beyond empirical risk minimization. In Bengio, Y., & LeCun, Y. (Eds.), Proceedings of The 6th International Conference on Learning Representations - ICLR 2018, Vancouver, Canada, April 30-May 3 (1-13). https://doi.org/10.48550/arXiv.1710.09412.
Zhong, Z., Zheng, L., Kang, G., Li, S., & Yang, Y. (2017). Random erasing data augmentation. In Proceedings of The AAAI Conference on Artificial Intelligence (13001-13008). https://doi.org/10.1609/aaai.v34i07.7000.