SISTEM PENGENALAN WAJAH UNTUK APLIKASI PRESENSI SISWA
The manual presence system currently applied in the schools is deemed inadequate to prevent the students from being absent, as this system fails to provide real-time report to schools of students' parents. Therefore, it is necessary to establish a quick, accurate and real-time presence system i...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Theses and Dissertations NonPeerReviewed |
| Published: |
[Yogyakarta] : Universitas Gadjah Mada
2011
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/91023/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=53818 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universitas Gadjah Mada |
| Summary: | The manual presence system currently applied in the schools is deemed
inadequate to prevent the students from being absent, as this system fails to
provide real-time report to schools of students' parents. Therefore, it is necessary
to establish a quick, accurate and real-time presence system in reporting to the
schools and students� parents when a student is absent. The use of barcode and
fingerprint scanning-based presence system has, however, some disadvantages,
thus requiring alternative method in order that the system can accurately
recognize the user�s identity. Such method may be of presence system using face
recognition.
Face image owned by specific person has of course a unique
characteristic, which makes the face recognizable. The face recognition method
applied in this research is eigenface method. The size of face image used in this
research is 80 x 80 pixels. This size is deemed efficient with the quick computation
time consumption and excellent accuracy. Too small image size will accelerate
computation time but reducing rate of accuracy, and vice verse. The image from
webcam first detects the face position and continued with cropping process. The
result of cropping is resized into 80 x 80 pixels. The resulting image is then
inputted into face recognition application. Student�s presence recapitulation is
saved in database accessible via website. As the result, the schools and student�s
parents may access it everywhere in real-rime.
Based on the research result, the conclusion can be drawn that eigenfacebased
face recognition method may be used in student�s presence application.
One of the factors to improve the system's accuracy is multiplying training images
for every individual. In this research, the researcher has conducted modification
in calculation method of eigenface if compared with the previous research. The
calculation of absolute rate of eigenface has successfully achieved rate of
accuracy 97.9%. Such result is 22.9% higher that of the previous research. This
eigenface-based face recognition application is highly sensitive to change in scale
and movement of face translation on the tested image, but not to rotational
movement of face and direction of moderate image capturing. |
|---|