FREDIANTO, - (2021) SISTEM IDENTIFIKASI TUMBUHAN OBAT BERDASAR POLA DAUN MENGGUNAKAN INVARIAN MOMEN HU DAN K-NEAREST NEIGHBOR. Tugas Akhir thesis, University of Technology Yogyakarta.
|
Text
5170411101_FREDIANTO_ABSTRAK.pdf Download (102kB) | Preview |
Abstract
FREDIANTO Department of Informatics, Faculty of Science & Technology University of Technology Yogyakarta North Ringroad St., Jombor Sleman Yogyakarta Email: fredianto248@gmail.com ABSTRACT Medicinal plants are plants that have the benefits to prevent or cure various diseases. The number of medicinal plants and our lack of knowledge about medicinal plants makes it difficult for people to distinguish between these types of medicinal plants. So that, many people prefer to use chemical drugs than medicinal plants. Therefore, in this study, the authors built a medicinal plant identification system. This study aims to build a medicinal plant identification system based on leaf patterns using hu moment invariance and k-nearest neighbour with Euclidean distance, manhattan distance, and Chebyshev distance. In this study, the process of identifying leaf images begins with the cropping and resizing processes. The feature extraction stage uses hu moment invariance. The final stage of classification uses a k-nearest neighbour. This research shows that the system can identify medicinal plant species well at the value of k = 3 with the euclidean distance and Chebyshev distance calculation algorithm. With the highest average accuracy for all types of leaves at 86,67%. Then the result was obtained an average accuracy of 72,22% on avocado leaves. For guava leaves, the average accuracy is 79,63%. Next, papaya leaves have an average accuracy of 65,74%. As in bay leaves, the average accuracy is 62,04%. The latter, betel leaf, has an average accuracy of 100%. Keywords: Identification System, Medicinal Plants, Invarian Momen Hu, K-Nearest Neighbor
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Informatika |
| Depositing User: | Kaprodi S1 Informatika UTY |
| Date Deposited: | 07 Sep 2021 04:33 |
| Last Modified: | 07 Sep 2021 04:33 |
| URI: | http://eprints.uty.ac.id/id/eprint/7994 |
Actions (login required)
 |
View Item |
