| Deep Learning-Based Classification of Date Palm Leaf Health: A Comparative Study of Convolutional Neural Network Architectures |
| Paper ID : 1001-ICEEM2025 (R1) |
| Authors |
|
El-Sayed M. El-Kenawy *1, Khaled Sh. Gaber2, Mahmoud Elshabrawy Mohamed3, Amel M Ali Alhussan4, Marwa M. Eid5 1Department of Communications and Electronics, Delta Higher Institute of Engineering and Technology, Mansoura 35111, Egypt 2Computer Science and Intelligent Systems Research Center, Blacksburg 24060, Virginia, USA 3Computer Engineering and Control Systems Department, Faculty of Engineering, Mansoura University, Egypt 4Department of Computer Sciences, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh 11671, Saudi Arabia 5Faculty of Artificial Intelligence, Delta University for Science and Technology, Mansoura 11152, Egypt |
| Abstract |
| Diagnosis of plant diseases, particularly in crops such as Date Palm, is crucial in conserving agricultural production gains. This study tackles the challenge of accurately identifying the health status of Date Palm leaves by using deep learning models to classify leaves as healthy or diseased, specifically focusing on two common conditions: the diseases include brown spot disease and white scale infection. Given data involving 750 images of Date Palms categorized into the three classes of healthy, brown spot and white scale disease, three CNN-based models were trained, namely GoogLeNet, AlexNet and VGG19Net, to recognize the characteristic labels given with Date Palm leaves. Measures of model performance are accuracy, sensitivity, specificity, F measure, Positive Predictive Value and Negative Predictive Value. According to the results, the proposed GoogLeNet model was the most accurate, with the highest accuracy of 95.14% of correct classification and the best specificity for classifying diseases from healthy plant leaves. While recognizing the capability of deep learning models, the current paper seeks to contribute to precision agriculture by identifying a method to diagnose diseases of Date Palms with a high degree of precision through automation. |
| Keywords |
| Date Palm, Deep Learning, Plant Disease Detection, CNN Models, GoogLeNet, Agricultural Health |
| Status: Accepted |