Related Experiment Video
Updated: Nov 3, 2025
08:20
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
2.0K
Identification of Cotton Leaf Lesions Using Deep Learning Techniques
Rafael Faria Caldeira1,2, Wesley Esdras Santiago2, Barbara Teruel1
1Faculty of Agricultural Engineering of the University of Campinas, FEAGRI/UNICAMP, Campinas 13083-875, Brazil.
Sensors (Basel, Switzerland)
|June 2, 2021
Summary
Deep learning models accurately identify cotton leaf lesions for improved crop health monitoring. Convolutional neural networks offer a more precise and reliable method for early pest and disease detection in cotton farming.
Area of Science:
- Agricultural Science
- Computer Science
- Plant Pathology
Background:
- Cotton is a globally significant agricultural crop, vulnerable to numerous pests and diseases in tropical regions.
- Early-stage symptoms of cotton diseases are often indistinguishable, complicating timely producer intervention.
- Accurate identification of leaf lesions is crucial for effective cotton crop management.
Purpose of the Study:
- To develop and evaluate deep learning models for automated identification of cotton leaf lesions.
- To enhance the monitoring of cotton crop health and support data-driven management decisions.
- To compare the efficacy of deep learning approaches against traditional image processing techniques.
Main Methods:
- Utilized deep learning, specifically convolutional neural networks (CNNs), with GoogleNet and Resnet50 architectures.
- Trained models on field images of cotton crops to detect and classify leaf lesions.
- Benchmarked CNN performance against traditional methods like Support Vector Machines (SVM), k-Nearest Neighbors (KNN), Artificial Neural Networks (ANN), and Neuro-Fuzzy Control (NFC).
Main Results:
- Resnet50 achieved a precision of 89.2%, and GoogleNet achieved 86.6% in identifying cotton leaf lesions.
- CNNs demonstrated up to 25% greater precision compared to SVM, KNN, ANN, and NFC methods.
- The proposed deep learning solution enables more rapid and reliable inspection of cotton plants.
Conclusions:
- Deep learning models, particularly CNNs, provide a highly accurate and efficient solution for identifying cotton leaf diseases.
- This technology can significantly aid producers in early detection and management of crop health issues.
- The findings suggest a promising advancement in automated agricultural monitoring systems.
