PENDETEKSIAN WAJAH SECARA REALTIME MENERAPKAN JARINGAN BACKPROPAGATION

Cut Lika Mestika Sandy

Abstract

Face detection is the first step required for various applications for computer vision, biometric recognition, and various multimedia applications, such as face tracking, face recognition, and video surveillance. In this research, a rapid face detection system was proposed applying the Backpropagation network. The system is developed from image qualification with the implementation of the sobel operator, and then the conditioned image is trained and tested. The performance of the developed system is considered based on two performance comparisons, detection rate and false positive rate. The assessment results show that the system performance that occurs has a detection rate ranging from 70% to 85%. For the development of system performance quality in future research, the developed system can be combined with one of the video complexity tests, skin color segmentation, or statistical tests.