Face Recognition in Kindergarten Students using the Principal Component Analysis Algorithm

In kindergarten, it is important to know the activities of each student to evaluate how individual student learns and adapts to school environment. Manually tracking individual student activities during class is hard for kindergarten teachers. In this paper, we propose face recognition in the kindergarten students as a first step to trace and record individual student activities. A video of kindergarten students is converted into digital images. Faces are detected using Viola-Jones method. The extraction feature on the images is done using the Principle Component Analysis (PCA) method. We implement Euclidean Distance to recognize student's face. Our experiments use 70 images as data training. The data training consists of 5 different images from 14 students. The experiment results show 91.42% accuracy by testing 14 new images of the students.