Opencv Python

Design and implement an online exam automatic proctoring system. For taking the exam and to avoid the student
having to install anything on their machines and running into technical problems, the university ships its students an
RPi-powered exam proctoring device. The device consists of a camera, the RPi, a single button, and several LEDs.

At the time of the exam, the student presses on the button to start the setup process. First, the take a picture LED
lights up. The user then looks at the camera and presses on the button.
Next, the take a picture of the id led lights up.
The student places the ID in front of the camera and presses on the button. The camera takes a picture of the id, reads
the id value using ocr, and compares it to the number preconfigured for the student (stored on his or her device).
If
the two match (id burned in the system and id card number), the next led lights up, which is for taking a video of the
environment. The student moves the easy to lift device to show the environment and when done presses the next button.
The system then lights an LED for positioning the device to show the face.
When the face is successfully detected, the
last LED is lit up to prompt the user to press the button one last time to begin the exam. A LED indicating that the
system is filming is lit up. When the exam ends.
Pressing the button for the last time will then turn off the filming LED
and send the video and images to the host computer (server). System overview is shown in Figure 1.
At the end of the day, the host computer sends an email to the faculty to share with them a report about all the
processed videos.
The host processes the video as follows:
1. It shows the total % of times a single face is detected
2. It shows the total % of times no face is detected
3.
It shows the total % of times multiple faces are detected.
If the percentage of no face detected and multiple faces detected combined are less than 5%, the system deletes the video.
If not, the system keeps the videos for faculty to check.

To keep the host from being overwhelmed and to maintain privacy, the host every day removes all videos and images
older than 1 week.