Title: Research on video image face detection and recognition technology based on improved MTCNN algorithm

Authors: Jinfeng Liu

Addresses: College of Robotics, Guangdong Polytechnic of Science and Technology, Zhuhai, Guangdong Province, China

Abstract: With the development of modern computer technology and artificial intelligence, face image processing technology has been widely used in people's life and work. In order to realise face image detection and recognition in dynamic video, this paper proposes a face detection and recognition technology based on MTCNN algorithm. MTCNN algorithm includes R-Net, O-net and P-net deep network models, which can realise face image deep processing in dynamic video. In order to train MTCNN algorithm deeply, Wider_Face and CelebA database training sets were used to train the additional test tasks and regression key points of the model. After setting the main parameters of MTCNN algorithm, the algorithm is simulated and analysed. Through the comparative simulation analysis of traditional algorithm, SVM algorithm and 2DPCA algorithm, it can be seen that MTCNN algorithm has more excellent performance and can meet the needs of face image detection and recognition in dynamic video.

Keywords: video image; MTCNN; face detection; distinguish; sample training.

DOI: 10.1504/IJWMC.2022.124811

International Journal of Wireless and Mobile Computing, 2022 Vol.22 No.3/4, pp.205 - 212

Received: 09 Oct 2021
Accepted: 12 Jan 2022

Published online: 09 Aug 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article