Multi-Stage Classification of Pomegranate Anthracnose Disease Severity Levels with CNN and SVM

• Published in: 2023 4th International Conference on Smart Electronics and Communication (ICOSEC) pp. 1117 - 1121
• Main Authors: Thakur, Sahil, Malik, Dhruv, Kukreja, Vinay, Sharma, Rishabh, Yadav, Rishika, Joshi, Kireet
• Format: Conference Proceeding
• Language: English
• Published: IEEE 20.09.2023
• Subjects: Agriculture; Anthracnose disease; Convolutional Neural Network; Convolutional neural networks; Crops; Electric potential; Machine learning; Neural networks; Pomegranate; Severity levels; Support vector machines; Task analysis
• DOI: 10.1109/ICOSEC58147.2023.10276047

This study applies a Convolutional Neural Network (CNN) model along with Support Vector Machines (SVM) to categorize the degrees of anthracnose disease severity. Various crops are susceptible to the fungal disease anthracnose, which results in severe damage and decreased yield. 88.93% overall accuracy was attained by the combined CNN-SVM model, according to the experimental results. The macro average values of precision, recall, and F1-scores for each class show satisfactory performance, and they are 53.27% on average. Overall accuracy across all classes is shown by the weighted average values to be 88.93%, and the same accuracy is also shown by the micro average values. These results demonstrate the CNN-SVM model's capability for various classification tasks and its accuracy in classifying the provided dataset. The research paper offers insightful information on the application of cutting-edge machine learning methods for identifying agricultural diseases, which can aid in the advancement of sustainable agricultural practices.