Classification of Various Diseases in the Mango Crop Using Machine Learning

Panuganti, Kavyasree (2024) Classification of Various Diseases in the Mango Crop Using Machine Learning. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (782kB) | Preview

Abstract

Mangoes are one among the most popular fruits which are widely popular for its taste and the nutritional values. Because of this huge demand mangoes are cultivated in vast areas across the world however these plants are affected by diseases due to pests and insects which will affect the quantity and the quality of the mangoes produced. Traditionally they are detected by checking them manually, which takes lot of time due to the manual labour work . With the rise in global demand for mangoes, the need for a fast, efficient, and automatic disease detection system has become critical. To achieve this the research is done to develop a robust hybrid framework leveraging deep learning techniques, such as Convolutional Neural Networks (CNN) and transfer learning, combined with traditional machine learning classifiers like Random Forest, Support Vector Machines (SVM), and XGBoost. The proposed approach in this research uses the pre-trained CNN architectures, including VGG16 (Visual Geometry Group with 16 layers) and VGG19(Visual Geometry Group with 19 layers) for feature extraction. K-means clustering is employed for doing the image segmentation to segment the mango leaf images into different regions based on pixel intensities. After extracting the features from CNN these are passed to traditional classifiers for disease classification which are evaluated on a mango leaf dataset containing 4000 images. The hybrid approach has performed well when compared to standalone CNN models. Among all the models Random Forest using VGG16 features has delivered the highest validation accuracy of 97.75%, while SVM and XGBoost models gave the competitive results with accuracies of 96.5% and 95.75%.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Niculescu, Hamilton UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science S Agriculture > SB Plant culture Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	04 Sep 2025 08:36
Last Modified:	04 Sep 2025 08:36
URI:	https://norma.ncirl.ie/id/eprint/8766

Actions (login required)

View Item