• Platform supported Windows, Linux, Mac
• Usable in Raspberry Pi

To create a face recognition that helps to watch specific places of the house was used Opencv. The idea is to build a house system that can match the face of the residents to give personal messages. The system sends a mail with the face and name (if is registered) of the person that was recognized.

The project face_recog_prj was built in first place as a system that runs over the computer or a Raspberry Pi, using a camera to recognizes faces.

For faces that aren't saved in "/Images", this tool will assign a name (actual date and hour), after, it will be saved in the "/Images" file. When that happened (even if is already saved in the file) the system will send a photo to the e-mail registered in "mail.txt"

1. Open your console and clone the repo:

git clone https://github.com/jumorap/face_recog_prj

2. Change the working directory to opera-extension-generator:

cd face_recog_prj

3. Install the requirements (You need to have installed Python3):

python3 -m pip install -r requirements.txt

1. Run i your console:

python3 Attendance.py

Note:
If do you wanna receive an e-mail when a face is recognized or added, you should edit the file: mail.txt and complete it with the instructions that appear there [email protected] your_password [email protected]

If you are in Windows, using python3 might not work. Use python instead.

• ModuleNotFoundError: No module named 'cv2' Then run in your console:

pip3 install opencv-python==4.5.2.52 --force --user

• ModuleNotFoundError: No module named 'face_recognition' Then run in your console:

pip3 install boost --force --user
pip3 install dlib==19.22.0 --force --user
pip3 install cmake==3.20.2 --force --user
pip3 install face_recognition==1.2.3 --force --user

Don't work? Download Conda environment here and run in your console:

conda install -c conda-forge dlib
conda install -c conda-forge cmake
conda install -c conda-forge face_recognition

Don't work?

1. Download Microsoft Visual Studio 2015 or newer (check if build tools are enough).
2. Install the dependencies from the previous step, using pip

MIT © Face Recog Prj