Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/113626| Title: | Biometrik Wajah Bagian Atas Monyet Ekor Panjang (Macaca fascicularis) sebagai Basis Data Sistem Pengenalan Temperamen Terotomatisasi |
| Other Titles: | Biometric Study of Cynomolgus Macaque's Upper Face as the Basis for Development of Automated Temperament Recognition System |
| Authors: | Esfandiari, Anita Dewi, Fitriya Nur Annisa Wahono, Jasmine Electra Alicia |
| Issue Date: | 2022 |
| Publisher: | IPB University |
| Abstract: | Penerapan prinsip refinement untuk meningkatkan kesejahteraan monyet
ekor panjang (MEP) sebagai salah satu hewan coba dapat dilakukan dengan
melihat kestabilan koloni melalui penilaian temperamen individu serta mengamati
keadaan klinis melalui ekspresi wajah. Perangkat Automated Face Recognition
(AFR) dapat digunakan untuk mengidentifikasi ekspresi satwa sebagai indikator
penentu temperamen. Dibutuhkan basis data berupa hasil evaluasi biometrik
wajah bagian atas untuk mengoperasikan perangkat tersebut. Hasil evaluasi akan
menunjukkan unit aksi wajah bagian atas yang dapat menjadi parameter penentu
ekspresi, termasuk yang berkorelasi dengan pain recognition. Penelitian dilakukan
dengan mengevaluasi pengukuran regio Action Unit (AU) secara numerik pada
beberapa foto. Hasil penelitian menunjukkan bahwa evaluasi numerik terhadap
AU1+2 (Brow raiser) yang diidentifikasi dengan mengukur luas daerah
underbrow dapat dijadikan parameter ekspresi aggressive pada MEP. Nilai
AU1+2 relatif tidak berbeda pada ekspresi lainnya sehingga diperlukan biometrik
wajah bagian bawah sebagai data komplementer. Evaluasi AU6 (Cheek raiser)
yang diidentifikasi melalui luas segitiga infraorbital, serta AU41 (Glabella
lowerer) yang diidentifikasi melalui jarak antara medial alis ke medial mata tidak
memberikan hasil numerik yang nyata untuk mengidentifikasi ekspresi sehingga
tidak dapat dijadikan parameter penentu. Identifikasi ekspresi melalui evaluasi
AU6 dan AU41 harus dilakukan bersamaan dengan evaluasi biometrik wajah
bagian bawah agar dapat mengidentifikasi suatu ekspresi. The implementation of refinement principles to increase laboratory cynomolgus macaque’s welfare can be done by observing social stability based on the temperament of each individual and observing clinical conditions through facial expressions. The development of Automated Face Recognition (AFR) device for nonhuman primates is currently underway with the aim to refine the identification of various facial expressions as indicator of temperament and welfare. The aim of this study was to evaluate biometrics of the upper face to identify the facial action unit that numerically changes with different expression, specifically those that correlate with pain recognition and stress condition. The study was conducted by quantifying the facial Action Unit (AU) of the upper face regions utilizing digital images obtained through observation of cynomolgus macaques that exhibited aggressive, neutral, affiliative, and fearful expressions. The results showed that measurement of AU1+2 (Brow raiser), which was identified by measuring the underbrow area, can be used as a parameter to identify aggressive expression. The evaluation of AU6 (Cheek raiser), which was identified through the area of the infraorbital triangle, and AU41 (Glabella lowerer) which was identified through the distance between the medial eyebrows to the medial eye, did not differ between different expressions which suggest that these AUs may not be useful as determinant parameter for temperamen recognition when used as a single unit. Expression identification by evaluation of AU6 and AU41 should be performed alongside evaluation of lower facial biometric. |
| URI: | http://repository.ipb.ac.id/handle/123456789/113626 |
| Appears in Collections: | UT - Veterinary Clinic Reproduction and Pathology |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| FULL TEXT.pdf Restricted Access | Full Text | 1.24 MB | Adobe PDF | View/Open |
| COVER.pdf Restricted Access | Cover | 435.17 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.