Author(s):
1. Shukria Muslim:
Institute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture,Peshawar,Pakistan
2. Naila Irfan:
Institute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture,Peshawar,Pakistan
3. Naveed Ullah:
Institute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture,Peshawar,Pakistan
4. Imran Uddin:
Institute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture,Peshawar,Pakistan Imran Uddin
Abstract:
Agriculture plays a vital role in the economy of the Khyber Pakhtunkhwa province in Pakistan, contributing over 20% to the provincial GDP and employing more than 40% of the labor force. Tomato cultivation is a major cash crop, yet farmers face challenges due to various tomato diseases, impacting yield and quality. This research proposes a novel approach for tomato disease detection using a lightweight and highly accurate Convolutional Neural Network (CNN) model optimized for resource-constrained environments. The model employs a hybrid CNN architecture integrating depth wise separable convolutions and attention mechanisms, initially trained on large-scale Plant Village datasets with transfer learning. Progressive knowledge distillation from pre-trained teacher models is applied to fine-tune the model, ensuring high accuracy while minimizing resource requirements. To enable deployment on lowpower devices such as smartphones or edge computing platforms, the model is compressed through pruning, quantization, and automated compression techniques. This efficient, robust solution enhances sustainable agricultural practices, supporting farmers in resource-limited settings and contributing to improved crop management and global food security.
Page(s):
1-1
DOI:
DOI not available
Published:
Journal: Second International Conference on Computing Technologies, Tools and Applications (ICTAPP-24), June 4-6,2024 (Abstract Book), Volume: 0, Issue: 0, Year: 2024
Keywords:
deep learning
,
Artificial Intelligence
,
Convolutional Neural Network CNN
,
Tomato Disease Detection
References:
References are not available for this document.
Citations
Citations are not available for this document.