Home security system based on face recognition techniques

• Main Author: Yuan, Wanheng
• Other Authors: Soong Boon Hee
• Format: Theses and Dissertations
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/78566
• Institution: Nanyang Technological University

Face recognition belongs to the category of computer vision. It refers to a smart technology that computers use to analyze and compare facial visual features to automatically identify themselves. In applications of image understanding and analysis, face recognition is very important and it is a great attempt in the process of developing biological features. Face recognition uses a computer to do analysis for the face video or image and extract the effective personal identification information from it, finally determine the identity of the face object. In this dissertation, I use two different approaches to do the face recognition. The first one is using a pre-trained network to embed and recognize face of my dataset. I use the library in OpenCV to do the detection and recognition. The system can detect, recognize people and distinguish people that haven’t been embedded in the system. In another approach, I constructed the convolutional neural network using Pytorch and use my own dataset to train it. Then I can use this to do recognition and test the accuracy of the system using the testing dataset.