Biometric authentication with low cost microcontroller

• Main Author: Tan, Sin Hui
• Other Authors: Mohammed Yakoob Siyal
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition
• Online Access: http://hdl.handle.net/10356/73009
• Institution: Nanyang Technological University

The purpose of this report is to elaborate and build a low cost facial identification system based on microcontroller as a part of the security feature. The facial identification method in this project is for facial recognition and drowsiness detection. Facial identification serves as a security tool to check the status of a person while using the hardware or software application. This method is popular for attendance-taking system, driving system, iPhone X etc. There are quite a few popular software tools to carry out the facial identification such as OpenCV, Kairos, Amazon, Microsoft etc. Each of the software has their own advantages and disadvantages. For this project we are using the OpenCV to build the facial identification system due to its available resources and it has the potential to expand in future as it support multi-platform. The facial identification in this project is built on the low-cost controller Raspberry Pi 3. The capability and reliability of Raspberry Pi 3 to handle the facial identification through OpenCV enable the author to successfully deploy it in this the project. A series of experiments is carried out to identify the factors affecting the performance of facial identification with Raspberry Pi and OpenCV. Based on the results, the overall performance looks promising in security control area.