Bhargav Rajyagor,
Paresh Vora,
- Assistant Professor, Department of Computer Application, Noble University, Bamangam, Gujarat, India
- Assistant Professor, Department of Computer Application, Noble University, Bamangam, Gujarat, India
Abstract
Weeds present a major challenge to crop productivity by competing with crops for vital resources, including water, sunlight, and nutrients, often resulting in significant yield reductions. On a global scale, weeds are responsible for approximately 13.2% of annual crop losses, a quantity sufficient to feed nearly one billion people. These invasive plants disrupt agricultural systems and adversely impact crop yields. Given their uneven distribution in fields, ground or aerial robots are commonly used for targeted herbicide application, utilizing computer vision algorithms to detect weeds before treatment. In cotton cultivation, prevalent weeds such as horse purslane (Trianthema portulacastrum L.) and purple nutsedge (Cyperus rotundus L.) pose significant threats. Despite increasing interest in deep learning-based weed detection, progress remains limited due to the scarcity of extensive datasets. To address this challenge, we introduce CottonWeeds, a curated dataset featuring 7,000 images of horse purslane and purple nutsedge captured under varied conditions in Indian cotton fields. This dataset is designed to facilitate the development of real-time weed recognition models. Weed control strategies typically involve physical, mechanical, biological, and chemical methods. However, recent advancements in Unmanned Aerial Vehicles (UAVs) and deep learning technologies, particularly Convolutional Neural Networks (CNNs), have enabled precise weed detection, reducing costs and promoting sustainable agricultural practices. This study explores the integration of UAVs and CNNs to improve weed detection accuracy, address herbicide-resistant species, and advance sustainable farming solutions. Future research aims to enhance these approaches by incorporating advanced image processing and deep learning algorithms for automated feature extraction, thus tackling complex challenges in cotton weed management. Model performance will be evaluated using metrics such as precision, recall, and F1 score to ensure a comprehensive assessment and minimize false positives and false negatives. This version maintains the original intent and detail while rephrasing for uniqueness and improved readability.
Keywords: Weeds, Computer Vision, Deep learning, UAVs (Unmanned Aerial Vehicles)
[This article belongs to Journal of Aerospace Engineering & Technology ]
Bhargav Rajyagor, Paresh Vora. Sustainable Cotton Crop Productivity through Precision Weed Detection: A Deep Learning-Based Approach with UAV Integration. Journal of Aerospace Engineering & Technology. 2025; 15(01):19-26.
Bhargav Rajyagor, Paresh Vora. Sustainable Cotton Crop Productivity through Precision Weed Detection: A Deep Learning-Based Approach with UAV Integration. Journal of Aerospace Engineering & Technology. 2025; 15(01):19-26. Available from: https://journals.stmjournals.com/joaet/article=2025/view=197425
References
1. Wang P, Tang Y, Luo F, Wang L, Li C, Niu Q, Li H. Weed25: A deep learning dataset for weed identification. Front Plant Sci. 2022; 13, 1053329. https://doi.org/10.3389/fpls.2022.1053329
2. Mensah B, Rai N, Betitame K, Sun X. Advances in weed identification using hyperspectral imaging: A comprehensive review of platform sensors and deep learning techniques. J Agric Food Res. 2024; 18, 101388. https://doi.org/10.1016/j.jafr.2024.101388
3. Luo T, Zhao J, Gu Y, Zhang S, Qiao X, Tian W, Han Y. Classification of weed seeds based on visual images and deep learning. Inf Process Agric. 2023; 10(1), 40–51. https://doi.org/10.1016/j.inpa.2021.10.002
4. Rajyagor B, Rakholia R, Agrawal SS. Tri-level handwritten text segmentation techniques for Gujarati language. Indian J Sci Technol. 2021; 14(7), 618–627. https://doi.org/10.17485/IJST/v14i7.2146
5. Karim Md. J, Nahiduzzaman Md, Ahsan M, Haider J. Development of an early detection and automatic targeting system for cotton weeds using an improved lightweight YOLOv8 architecture on an edge device. Knowl-Based Syst. 2024; 300, 112204.
https://doi.org/10.1016/j.knosys.2024.112204
6. Rai N, Mahecha MV, Christensen A, Quanbeck J, Zhang Y, Howatt K, Ostlie M, Sun X. Multiformat open source weed image dataset for real-time weed identification in precision agriculture. Data Br. 2023; 51, 109691. https://doi.org/10.1016/j.dib.2023.109691
7. Coleman GRY, Kutugata M, Walsh MJ, Bagavathiannan MV. Multi-growth stage plant recognition: A case study of Palmer amaranth (Amaranthus palmeri) in cotton (Gossypium hirsutum). Comput Electron Agric. 2024; 217, 108622.
https://doi.org/10.1016/j.compag.2024.108622
8. Rizvi SMH, Naseer A, Rehman SU, Akram S, Gruhn V. Revolutionizing agriculture: machine and deep learning solutions for enhanced crop quality and weed control. IEEE Access. 2024;12:11865–78. doi: 10.1109/ACCESS.2024.3355017.
9. Qu HR, Su WH. Deep learning-based weed–crop recognition for smart agricultural equipment: a review. Agronomy. 2024;14(2):363. doi: 10.3390/agronomy14020363.
10. Fan X, Chai X, Zhou J, Sun T. Deep learning based weed detection and target spraying robot system
at seedling stage of cotton field. Comput Electron Agric. 2023;214:108317. doi: 10.1016/j.compag.2023.108317.
11. Kakade S, Pawar P, Anekar D, Patil Y, Rathod S, Shah F. A study of weed management in agricultural plants using IoT and machine learning. 2024 International Conference on Emerging Smart Computing and Informatics (ESCI), Pune, India. 2024. pp. 1-6. p. 1–6. doi: 10.1109/ESCI59607.2024.10497452
Journal of Aerospace Engineering & Technology
| Volume | 15 |
| Issue | 01 |
| Received | 20/01/2025 |
| Accepted | 25/01/2025 |
| Published | 30/01/2025 |
| Publication Time | 10 Days |
Login
PlumX Metrics