IJITEE (International Journal of Information Technology and Electrical Engineering)
Vol 3, No 1 (2019): March 2019

Prototype of Student Attendance Application Based on Face Recognition Using Eigenface Algorithm

Tio Eko Prabowo (Universitas Gadjah Mada)
Rudy Hartanto (Universitas Gadjah Mada)
Sunu Wibirama (Universitas Gadjah Mada)

Article Info

Publish Date
13 Sep 2019


Prototype of face recognition based attendance application that has been developed to overcome weaknesses in DTETI UGM student manual attendance system has several weaknesses. These weaknesses are a decrease in facial recognition accuracy when operating under conditions of varying environmental light intensity and in condition of face rotating towards z axis rotation centre. In addition, application prototype also does not yet have a database to store attendance results. In this paper, a new application prototype has been developed using Eigenface face detection and recognition algorithm and Haar-based Cascade Classifier. Meanwhile, to overcome prototype performance weaknesses of the previously developed application, a pre-processing method was proposed in another study was added. Processes in the method were geometry transformation, histogram levelling separately, image smoothing using bilateral filtering, and elliptical masking. The test results showed that in the category of various environmental light intensity conditions, face recognition accuracy from developed application prototypes was 16.71% better than previous application prototypes. Meanwhile, in category of face slope conditions at z axis rotation centre, face recognition accuracy from developed application prototype was 38.47% better. Attendance database system was also successfully implemented and running without error.

Copyrights © 2019

Journal Info





Electrical & Electronics Engineering


IJITEE (International Journal of Information Technology and Electrical Engineering), with registered number ISSN 2550-0554 (Online), is a peer-reviewed journal published four times a year (March, June, September, December) by Department of Electrical engineering and Information Technology, Faculty ...