Publication: Vision-based plant health monitoring using deep learning models
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering::Electrical and electronic engineering | |
| dc.contributor.author | Muhammad Amirul Ikhwan bin Shamsuri | |
| dc.date.accessioned | 2025-05-13T04:50:41Z | |
| dc.date.available | 2025-05-13T04:50:41Z | |
| dc.date.issued | 2023-07 | |
| dc.description.abstract | The horticulture industry in Malaysia is a significant contributor to the country's agricultural sector, encompassing the production and cultivation of a wide range of crops, including fruits, vegetables, flowers, and ornamental plants. This sector is diverse and dynamic, producing tropical fruits like rambutan, durian, guava and mango. However, this industry is not without its challenges, pests and diseases can adversely affect the health and productivity of plants. In an effort to tackle these challenges, a novel system has been proposed that leverages machine learning techniques to identify plant health through convolutional neural networks (CNN). Farmers can leverage the CNN model from this research as a blueprint and foundation to create their own customized plant health monitoring system that addresses the specific requirements and conditions of their crops and farm setting. By reducing the time and effort required to monitor plant health, it can increase agricultural productivity while using fewer resources. Hence, the identification of plant diseases using CNN algorithms will be studied by using the CNN models such as AlexNet, InceptionV3, MobileNetV3, ResNet50 and VGG19 to compare, analyse, and determine the best suit models to be used in identifying the diseased using deep learning technique. At the end of this research, ResNet50 has achieved the highest accuracy percentage that is 97.36% when compared to MobileNetV3 (96.70%), InceptionV3 (95.60%), VGG19 (94.20%) and AlexNet (87.47%). | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/21599 | |
| dc.language.iso | en | |
| dc.title | Vision-based plant health monitoring using deep learning models | |
| dc.type | Resource Types::text::report::technical report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |