swagatobag2000 / face-recognition-openCV

This module gets first trained by the pre-trained haar-cascade classifier, then using collected face data, it recognizes the user.

face-recognition

Screenshots

How to make face-recognition module

1. Install required Python Packages:

pip install opencv-python

2. Download the pre-trained classifier

• Download "haarcascade_frontalface_default.xml" from haarcascade

3. Start Testing Face Detection

• use "opencv-face-testing.ipynb" to check if openCV is working or not

# Choose an image to detect faces in
frame = cv2.imread('swagato.jpeg')

# Iterate forever over frames
while True:
    # Read the current frame
#     successful_frame_read, frame = webcam.read()
    # Must convert to grayscale
    grayscaled_img = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)
    ....

 

• We can see the static image is detected by green rectangle

# Iterate forever over frames
while True:
    # Read the current frame
    successful_frame_read, frame = webcam.read()
    # Must convert to grayscale
    grayscaled_img = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)
    ....

• now, uncomment the line which we commented in previous step, to fetch live image with webcam

• webcam application will pop up and one rectangle will show up when we run the notebook like above.
• using haarcascade classifier we are detecting faces
• if we press 'q', the webcam window will be closed.

4. Collect Data

• run "face-data-collect.ipynb" to collect data as .npy or numpy files
• It will first ask name of the person and the face data will be recorded as a numpy file followd by this
• file will be saved at "data" directory

5. Perform Face Recognition

• run "face-recognition" to test face recognition