FEATURE PYRAMID NETWORK BASED REAL-TIME YOLO FACIAL MASK RECOGNITION

FEATURE PYRAMID NETWORK BASED REAL-TIME YOLO FACIAL MASK RECOGNITION

Mrs. M.K. Geedtha ,
Dept of Computer Science and Engineering,

Guide

Mr. A.N. Arun B.E.,M.E.,M.B.A.,SCPM

Head of the department

Dept of Computer Science and Engineering

Sri Venkateshwara Insitute of Science and technology, Thiruvallur ,Tamilnadu

Abstract— In recent years, automated recognition of faces and facial features has grown in popularity and significance. Face changes and the presence of masks make this task extremely difficult. A large amount of research has been done to recognise faces under various conditions such as changing pose or illumination, degraded images, and so on. Despite this, recognising people's faces behind masks remains difficult. The primary goal of this work is to make the system detect facial masks and, in particular, to improve the recognition accuracy of various masked faces. The proposed method involves first detecting the facial regions. A Feature Pyramid Network (FPN) learner is used to solve the occluded face detection problem.

Keywords— Feature Pyramid Network (FPN), YOLO (You Only Look Once), Support Vector Machine (SVM)