Face Recognition Based on Viola-Jones Algorithm as Dataset for Image Classification

Face detection is the first step in the face recognition system to identify the entire image area to find facial areas and non-face areas. The face recognition process can be known through the position and distance of the face from the camera. The Viola-Jones and HOG algorithms have good accuracy in facial recognition. This is evident from the experimental results obtained that face recognition using this algorithm has a recognition ability of 98.9%. This algorithm can also be used to create a dataset from an image. In this study, image classification was performed using three models, that is Naive Bayes, random forest, and logistic regression. The dataset used comes from facial recognition performed with Viola-Jones and HOG. The result obtained from this face classification is that the logistic regression model has the highest level of accuracy because it can classify all images perfectly. The Naive Bayes model has one error in image classification, and the random forest model has three errors in image classification.