Related Experiment Video
Updated: Jan 7, 2026
08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
2.0K
Lightweight residual graph augmented transformer for cassava leaf disease recognition using spectral directional
T Satheesh1, M S Geetha Devasena2
1Department of Artificial Intelligence and Data Science, Dr.N.G.P. Institute of Technology, Coimbatore, 641048, India. cse.satheesh@gmail.com.
Scientific Reports
|December 30, 2025
Summary
A new model, Lite-RGA-GTNet, accurately detects cassava leaf diseases using advanced AI. This technology aids sustainable farming by improving crop yield and disease management in real-time.
Area of Science:
- Agricultural Science
- Computer Vision
- Machine Learning
Background:
- Cassava leaf disease detection is crucial for crop yield and food security.
- Challenges include similar symptoms, variable conditions, and computational demands of existing models.
Purpose of the Study:
- To develop an efficient and accurate deep learning model for cassava leaf disease detection.
- To address limitations of existing models in capturing fine textures and spatial relationships.
Main Methods:
- Introduced Lite-RGA-GTNet, a lightweight network integrating RGB, gradient, and vegetation index data.
- Employed spectral-directional preprocessing, residual graph reasoning, and progressive token pruning.
- Utilized hierarchical graph-transformer modules for fused local-global feature representation.
Main Results:
- Lite-RGA-GTNet achieved 96.84% accuracy, 96.25% precision, 96.72% recall, and 96.48% F1-score.
- Outperformed existing models like CassNet and LeafXFormer by up to 2.65% in accuracy.
- Demonstrated an average inference time of 14 ms for real-time deployment.
Conclusions:
- Lite-RGA-GTNet offers a highly accurate and computationally efficient solution for cassava leaf disease detection.
- The model's performance indicates its suitability for real-time agricultural applications and sustainable farming.
