Embedded system application development on raspberry Pi 3 - (B) : RazPy (face recognition based home surveillance system)

• Main Author: Woon, Yee Py
• Other Authors: Chong Yong Kim
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/75054
• Institution: Nanyang Technological University

Closed-circuit Television (CCTV) is widely used for security and surveillance purposes. Currently, the limitations of CCTVs include poor footage quality, low frame rate and high memory consumption. These limitations can be solved by incorporating computer vision technology such as face detection and face recognition into CCTV. RazPy is a face recognition-based home surveillance system which is implemented by Raspberry Pi 3 Model B, Python programming language and OpenCV library. By using Local Binary Pattern (LBP) algorithm, RazPy is able to perform face recognition accurately. Most importantly, RazPy can improve the quality of footage and reduce memory consumption by recording footage only when unknown face(s) are detected. Besides, additional features such as email alert and live stream are included in RazPy to provide real time information to users. At the end of this project, RazPy was able to perform face recognition accurately, send email to users, record footage and support live stream. Further research on deep learning algorithm can be done to further improve the accuracy face of recognition.