NORMA eResearch @NCI Library

Detect Foliar Disease in Apple Trees using Deep Learning

Said, Aishwarya Ishwar (2022) Detect Foliar Disease in Apple Trees using Deep Learning. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (4MB) | Preview


Apples are one of the most important fruits in the world. We all have heard the famous saying ’an apple a day keeps the doctor away’ which means that eating an apple a day helps us to maintain good health. Leaf disease is one of the major threats to the overall quality of an apple orchard. The traditional method of identifying the disease requires manual scouting of apple orchards by humans, which is time-consuming and expensive. This research focuses on exploring different deep learning approaches to identify leaf disease. In this study, we have used the standard convolution-neural network and transfer-learning based model to identify and diagnose leaf diseases in apple trees. Along with this, we would also explore different data augmentation techniques and analyse their effects on the overall accuracy of the model. All the implemented models are trained on the Plant Pathology 2021 dataset which consists of approximately 23,000 high-quality RGB images of apple leaves with 12 different foliar diseases. Out of 23,000, 18632 images are unhidden and are available to the Kaggle community to train machine learning models. However, due to computational limitations, we have used only 40% of the data such that the proportion of each disease remains the same.

Item Type: Thesis (Masters)
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: Tamara Malone
Date Deposited: 10 Mar 2023 16:13
Last Modified: 10 Mar 2023 16:13

Actions (login required)

View Item View Item