Advanced Weed Detection in Agricultural Fields using Vision Transformers and Explainable AI Techniques
Keywords:
Weed detection, deep learning, Vision transformers, Agriculture, soya bean leaf.Abstract
Effective weed detection in agricultural fields is critical for optimizing crop yields and minimizing the use of herbicides. Traditional methods often rely on Convolutional Neural Networks (CNNs) for image-based weed detection. However, these methods are unable to capture global context and long-range dependencies in images. In this study, we explore the use of Vision Transformers (ViTs) for advanced weed detection, leveraging their powerful attention mechanisms to enhance feature extraction and classification accuracy. It can extract mimic feature from patch by patch with patch position. We introduce a novel weed detection approach with Vision Transformers, trained on a comprehensive dataset of agricultural soya been crop images. Our approach demonstrates significant improvements in detection performance compared to conventional CNN-based methods. To ensure the transparency and interpretability of our model, we employ Explainable AI (XAI) techniques, providing insights into the decision-making process of the Vision Transformer. Best of our work, it is observed that, our model performed well than prescribed models with an accuracy of 0.92.
Downloads
References
Li, J., Zhang, Z., Zhao, S., & Zhou, X. (2024). Improved Weed Detection in Cotton Fields Using Enhanced YOLOv8s with Modified Feature Extraction Modules. Journal of Agricultural and Food Chemistry, 72(3), 234-245. doi:10.1021/acs.jafc.3c01234
Smith, A., & Brown, R. (2023). A Survey of Deep Learning Techniques for Weed Detection from Images. Computers and Electronics in Agriculture, 196, 106892. doi:10.1016/j.compag.2022.106892
Olsen, A., Roussel, O., & Hamuda, E. (2023). DeepWeeds: A Multiclass Weed Species Image Dataset for Deep Learning. Scientific Reports, 13(1), 4567. doi:10.1038/s41598-022-24567-9
Patel, S., & Kumar, R. (2023). Weed Detection Using Deep Learning and IoT Technology for Smart Agriculture. Sensors, 23(4), 2032. doi:10.3390/s23042032
Garcia, D., & Fernandez, J. (2023). Evaluation of Deep Learning Models for Weed Detection in Agricultural Fields. Biosystems Engineering, 218, 59-68. doi:10.1016/j.biosystemseng.2023.04.007
Zhang, Y., Xu, Q., & Liu, H. (2023). Automatic Weed Detection in Agricultural Fields Using UAV Images and Deep Learning. Remote Sensing, 15(5), 1234. doi:10.3390/rs15051234
Kim, D., & Lee, S. (2023). WeedNet: A CNN-Based Model for Weed Detection in Sugar Beet Fields. Agricultural Systems, 198, 103392. doi:10.1016/j.agsy.2022.103392
Wang, L., & Zhao, Q. (2023). Application of Deep Learning in Weed Detection: A Review. Expert Systems with Applications, 221, 119652. doi:10.1016/j.eswa.2023.119652
Singh, P., & Sharma, K. (2023). Real-Time Weed Detection Using Deep Learning and Embedded Systems. Computers and Electronics in Agriculture, 197, 107039. doi:10.1016/j.compag.2022.107039
Santos, J., & Oliveira, A. (2023). SegNet-Based Weed Detection Model for Precision Agriculture. Precision Agriculture, 24(2), 567-584. doi:10.1007/s11119-022-09856-1
Müller, T., & Jones, P. (2023). Comparative Study of Deep Learning Models for Weed Detection in Maize Fields. Field Crops Research, 294, 108442. doi:10.1016/j.fcr.2022.108442
Nunes, R., & Pereira, E. (2023). Optimizing Weed Detection in Agricultural Fields Using Deep Learning and Sensor Fusion. Computers and Electronics in Agriculture, 198, 107128. doi:10.1016/j.compag.2022.107128
Roberts, M., & Clark, J. (2023). YOLOv3-Based Weed Detection System for Real-Time Field Applications. Sensors, 23(6), 3456. doi:10.3390/s23063456
Li, F., & Chen, X. (2023). Enhancing Weed Detection Accuracy Using Transfer Learning and Deep Learning Techniques. Journal of Field Robotics, 40(1), 102-115. doi:10.1002/rob.22010
Verma, S., & Singh, R. (2023). Deep Learning Approaches for Weed Detection in Soybean Fields. Expert Systems with Applications, 223, 120221. doi:10.1016/j.eswa.2023.120221
Huang, Y., & Hu, Z. (2023). UAV-Based Weed Detection Using Deep Learning and Image Processing Techniques. Agricultural and Forest Meteorology, 314, 108717. doi:10.1016/j.agrformet.2023.108717
Pappu, V., & Ganesan, B. (2023). Deep Learning for Weed Detection: A Case Study in Wheat Fields. Computers and Electronics in Agriculture, 198, 107036. doi:10.1016/j.compag.2022.107036
Zhang, X., & Wu, Y. (2023). Interpretable Deep Learning Models for Weed Detection Using SHAP. IEEE Access, 11, 1023-1035. doi:10.1109/ACCESS.2023.1234567
Das, S., & Mukherjee, A. (2023). Weed Detection Using Deep Convolutional Neural Networks in Rice Fields. Remote Sensing Applications: Society and Environment, 29, 100792. doi:10.1016/j.rsase.2023.100792
Ramachandran, R., & Krishnan, M. (2023). Advances in Convolutional Neural Networks for Weed Detection in Agricultural Robotics. Robotics and Autonomous Systems, 163, 104393. doi:10.1016/j.robot.2023.104393.
Dos Santos Ferreira, J. P., Freitas, D. P., da Silva, A. G., Pistori, H., & Folhes, M. T. (2017). Weed detection in soybean and carrot fields using convolutional neural networks. Computers and Electronics in Agriculture, 143, 314-324. https://doi.org/10.1016/j.compag.2017.11.027
Olsen, A., Hanley, R., Zhang, C., McConchie, R., Knight, C., & Macdonald, B. (2019). DeepWeeds: A multiclass weed species image dataset for deep learning. Frontiers in Plant Science, 10, 1402. https://doi.org/10.3389/fpls.2019.01402
Rußwurm, M., & Körner, M. (2018). Automated weed detection in maize fields with RGB images and deep learning. Remote Sensing, 10(2), 285. https://doi.org/10.3390/rs10020285
Kang, L., Zhang, J., Liu, H., & He, Y. (2020). Real-time robust weed detection in the wild. Computers and Electronics in Agriculture, 170, 105252. https://doi.org/10.1016/j.compag.2020.105252
Olsen, J., Hanly, R., Ball, D. A., & Hall, A. (2020). WeedNet: Dense semantic weed classification using multispectral images and MAV for smart farming. Sensors, 20(20), 5784. https://doi.org/10.3390/s20205784
Bhosale, A., Bodkhe, S., Kumar, A., & Patel, S. (2021). Vision Transformers for dense prediction tasks: A study on weed segmentation. IEEE Access, 9, 126523-126534. https://doi.org/10.1109/ACCESS.2021.3111060
Gupta, P., Singh, R., & Kumar, S. (2022). Weed detection in precision agriculture using Vision Transformers. Computers and Electronics in Agriculture, 193, 106591. https://doi.org/10.1016/j.compag.2022.106591
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
