Tan, Yin Ying (2015) Plant leaf classification using shape features and Bayesian Network. Universiti Malaysia Sabah. (Unpublished)
|
Text
ae0000002747.pdf Download (350kB) | Preview |
Abstract
Environment pollution and overwhelming development by human in recent decades has lead to rise in global warming which cause the plants are at risk of extinction. Therefore, it is very important to have a plant protection plan to manage and archive them from being forgotten by our future generation. With the rapid development of computer technology, plant classification has broken through the traditional methods, rapid identification for plant can be achieved by the image processing and pattern recognition technology with the leaf images provided. There are several features on the leaf that can be used to perform classification. Among of these features, shape is chosen for classification in the research presented in this report because the shape of a leaf is the most distinguishing feature compared to other features. The research presented proposed an approach to classify plant leaf using shape features extracted from the leaf shape images, which will then be forwarded as an input to classifier. A supervised method Bayesian Network that applied search and scoring metrics approach for structure learning is used as classifier. The experiments result is measured in term of average accuracy and standard deviation. The best accuracy obtained from the method proposed is 71.38%. Although the classifier proposed does not perform well, the proposed features are able to produce an accuracy of 83.56% with Neural Network classifier. This result outperformed most of the work in literature reviews that used different features. This shows that the features proposed is capable to produce better accuracy.
| Item Type: | Academic Exercise |
|---|---|
| Keyword: | plant protection plan, shape of a leaf, Bayesian Network, plant classification |
| Subjects: | ?? QA75 ?? T Technology > TD Environmental technology. Sanitary engineering |
| Department: | FACULTY > Faculty of Computing and Informatics |
| Depositing User: | ADMIN ADMIN |
| Date Deposited: | 09 Nov 2015 14:51 |
| Last Modified: | 27 Oct 2017 17:03 |
| URI: | https://eprints.ums.edu.my/id/eprint/12113 |
Actions (login required)
 |
View Item |