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
Adaptive preprocessing and Cascaded Canny Edge Segmentation for cassava disease identification using
1Department of Computer Science and Engineering, Excel Engineering College, Namakkal, Tamil Nadu, India.
Frontiers in Plant Science
|January 5, 2026
Summary
This study introduces an AI system for accurate cassava disease detection using HyperCapsInception-ResNet-V2-CNN and optimal feature selection. The system achieved high accuracy, improving precision farming and early disease identification.
Area of Science:
- Agricultural Science
- Computer Science
- Artificial Intelligence
Background:
- Cassava is a vital global crop, but identifying leaf diseases is challenging due to complex feature interdependencies.
- Previous methods suffered from high false positives and misidentification, reducing accuracy in disease detection.
Purpose of the Study:
- To develop an efficient AI-powered image analysis system for enhanced cassava disease detection.
- To improve the accuracy and precision of identifying disease regions in cassava leaves.
Main Methods:
- Utilized a dataset of 21,367 cassava leaf images from Kaggle.
- Employed adaptive Gaussian Otsu thresholding, histogram color evaluation, and iterative clustering for data normalization and segmentation.
- Applied Cascaded Canny Edge Segmentation (CCES) and Optimal Spider Swarm Intelligence Technique (OSSIT) for feature selection and dimension reduction.
- Classified diseases using a HyperCapsInception-ResNet-V2-CNN model.
Main Results:
- Achieved 98.15% accuracy, 97.22% F1-score, and 96.02% precision.
- Outperformed traditional methods including EfficientNetB3, AlexNet, Faster-RCNN, and InceptionV3.
- Demonstrated significant enhancement in detection and classification accuracy.
Conclusions:
- The proposed AI system effectively automates cassava disease detection.
- Optimized feature selection and the hybrid CNN architecture are key to improved accuracy.
- The system shows high potential for practical application in precision agriculture and early disease management.
