Abstract
Durian tree trunk diseases are threatening to reduce the productivity of farmers and therefore have economic consequences, such as lower yield and increased cost. Currently, diagnosis and treatment of diseases is a very mobile and subjective process, hence slow and not effective. In this project we do propose a construction of “DurianCare”, a mobile application that able to utilize techniques of Convolutional Neural Networks (CNN), Precision Agriculture, and Content Based Filtering for enhancing the farming practices of sustaining crops by advanced disease detection and the guided recommendation. The project will implement automated detection and disease management recommendation algorithms while evaluating the functionality and usability of the designed application. This could include for instance a systematic literature review on theories available to understand the current problem and its possible solution, analysis of the users need gathered from interviews, and of course the application development using Flutter within Visual Studio Code. The application includes advanced features such as automatic disease detection using a CNN model trained on a publicly available dataset, integration into precision farming (where environmental signals are monitored in real time), and content-based filtering for custom-tailored therapy proposals. During the application's development, a wide literature review was conducted, users were interviewed and surveyed, and the developed application was iteratively built and redesigned. Extensive functional and user testing was performed to ensure the application not only accomplishes its goal but increases efficacy as well. The mobile app will do early and accurate diagnosis of the diseases and suggestion of management practices which, in turn, can help to reduce the loss of farmers and ensure increased crop production using a limited resource with minimum loss to the farmers. To improve the welfare of farmers and the productivity of durian farming, this project focuses on development and promotion of proper and sustainable disease control methods.
Metadata
Item Type: | Thesis (Degree) |
---|---|
Creators: | Creators Email / ID Num. Azmi, Muhammad Haziq 2021820224 |
Contributors: | Contribution Name Email / ID Num. Thesis advisor Ismail @ Abdul Wahab, Zawawi UNSPECIFIED |
Subjects: | Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science > Neural networks (Computer science) |
Divisions: | Universiti Teknologi MARA, Terengganu > Kuala Terengganu Campus > Faculty of Computer and Mathematical Sciences |
Programme: | Bachelor of Computer Science (Hons) |
Keywords: | Durian Tree Trunk Diseases, Convolutional Neural Networks (CNN) |
Date: | 2025 |
URI: | https://ir.uitm.edu.my/id/eprint/115077 |
Download
![[thumbnail of 115077.pdf]](https://ir.uitm.edu.my/style/images/fileicons/text.png)
115077.pdf
Download (121kB)
Digital Copy
Physical Copy
ID Number
115077
Indexing

