Comparing MTCNN and Viola-Jones Algorithm in Face Recognition

Taking into account the importance of human face recognition in numerous applications such as identity authentication, determining gender and age, analyzing facial expressions, assessing head pose, evaluating lighting conditions, and overcoming obstacles, the aim of this study is to investigate and contrast the effectiveness of neural networks and the Viola-Jones algorithm in recognizing human faces. To achieve this specific goal, the Multi-Task Cascade Convolutional Neural Network (MTCNN) was used to evaluate the accuracy and processing time. In addition, the two algorithms MTCNN and Viola-Jones were compared in terms of accuracy and processing time. The results show that MTCNN can recognize the human face more accurately. At the same time, the processing time for face recognition with MTCNN is shorter than 71.63% compared to the processing time required by the Viola-Jones algorithm. The results clearly show that MTCNN works much faster, with a 71.63 % faster than the Viola-Jones algorithm.