Related Concept Videos
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Sort by
Same author
CA<sup>2</sup>PNet: a context-aware multi-scale architecture with adaptive attention and progressive dilated convolutions for biomedical image segmentation.
Frontiers in artificial intelligence·2026
Same author
An automated framework to classify skin lesions using Multi-Head Self Attention Layer-based Vision Transformers.
Frontiers in artificial intelligence·2026
Same author
Advanced IoT-integrated parking systems with automated license plate recognition and payment management.
Scientific reports·2025
Related Experiment Video
Updated: Feb 26, 2026

08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
2.1K
An attention-augmented lightweight convolutional framework for fine-grained plant leaf disease classification.
Adithiyaa D1, Lakshhmi Narayanan T1, Manas Ranjan Prusty2
1School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India.
Frontiers in Plant Science
|February 25, 2026
Summary
A new lightweight convolutional neural network (CNN), Attentive and Lightweight Network (ALNet), achieves high accuracy in plant leaf disease prediction. This efficient deep learning model significantly reduces parameters and model size for easier deployment on edge devices.
Area of Science:
- Computer Science
- Artificial Intelligence
- Machine Learning
Background:
- Deep learning models like CNNs and transformers are crucial for image classification.
- Transformers have shown impressive accuracy but often require substantial computational resources.
- There is a need for efficient deep learning models for deployment on resource-constrained devices.
Purpose of the Study:
- To propose a novel, lightweight CNN model named Attentive and Lightweight Network (ALNet).
- To achieve high classification accuracy for plant leaf diseases while minimizing model parameters and size.
- To facilitate the deployment of accurate disease prediction models on cloud platforms and edge devices.
Main Methods:
- Developed a custom lightweight CNN architecture (ALNet) comprising stem, core, and head blocks.
- The core classifier was inspired by established models like ResNet, SENet, EfficientNet, SqueezeNet, and ShuffleNet.
- Evaluated ALNet using 5-fold cross-validation on grapevine, apple, and cherry datasets.
Main Results:
- ALNet achieved 99.78% accuracy on multi-class grapevine classification and 100% on binary classification.
- Achieved 99.95% accuracy on multi-class apple classification and 100% on cherry binary classification.
- ALNet utilizes only 0.17 million parameters, is 18x smaller than SqueezeNet, requires 151.98 MFLOPs, and trains 1.2-2.2x faster per epoch.
Conclusions:
- ALNet demonstrates high accuracy in plant leaf disease prediction.
- The model's lightweight nature and reduced parameter count make it suitable for edge devices and cloud deployment.
- ALNet offers a compelling balance between performance and efficiency for agricultural applications.

