HAAR CASCADE CLASSIFIER DAN ALGORITMA ADABOOST UNTUK DETEKSI BANYAK WAJAH DALAM RUANG KELAS

Face detection is one of the important technological developments in computer vision field such as security systems, control systems, including the presence or attendance system. The proposed system leads to automatic attendance system but only to the extent that there is face detection student in the class based on the position of the face is straight ahead position , rotation parallel to right 15 o , 15 o rotation aligned to the left, lift the chin 15 o , 15 o head up and down and based on three distance face objects, in example 100 cm, 150 cm and 200 cm, and calculate the number of students that are in class. Based on this background we proposed multiple-face detection system in the classroom. The method used to detect faces using AdaBoost algorithms and applications written in the programming language C# in Visual Studio 2008. Face image taken from a webcam using EmguCV capture library functions. Tests carried out in the normal light intensity, in addition, the accuracy is also measured in terms of the real face are detected directly from the webcam. From the testing that has been done seven facial image results obtained with the three position and distance of the object can be detected with a good face of 8 face images in the classroom due to obstructed by other students. Keywords : AdaBoost algorithm, EmguCV, haar cascade classifier, multiple face detection