Kusumawati, Anisak (2020) IMPLEMENTASI LOCAL BINARY PATTERN DAN K-NEAREST NEIGHBOR UNTUK KLASIFIKASI CITRA WAJAH. Tugas Akhir thesis, University of Technology Yogyakarta.
|
Text
Naskah Publikasi-5160411245 Anisak Kusumawati.pdf Download (679kB) | Preview |
Abstract
There are many methods that can be used to recognize a person's identity, such as through the component it owns, an identity card, a unique number, a password. However there is a shortage of such methods such as identity cards can be lost, unique numbers and passwords can be forgotten. The solution to the problem is to identify a person based on biometrics such as face, Iris, and fingerprint, or from the combination of the three. In this study used the image of the face as identity recognition of a person's name. The extraction of features used are Local Binary Pattern, classification is done with the algorithm K-Nearest Neighbor (K-NN) and distance calculation Manhattan Distance. The results of the Nilak K test show that the highest accuracy lies at a value of k = 1 of 92,85% using point 24 and a radius of 9 with correct amounts of 39 data and an incorrect amount of data 3. The highest average accuracy lies at k = 1 of 77,21% and the lowest at k = 9 by 67,57%. Keywords: Face, Identity recognition, Local Binary Pattern, Manhattan Distance.
| 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: | 20 Dec 2020 02:14 |
| Last Modified: | 20 Dec 2020 02:14 |
| URI: | http://eprints.uty.ac.id/id/eprint/6342 |
Actions (login required)
 |
View Item |
