face_detection.py

import cv2
from random import randrange

# load pre-trained data on face frontals from opencv
trained_data = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')

# choosing image to detect faces in
img = cv2.imread('Cillian.png')


# capturing video from webcam
webcam = cv2.VideoCapture(0)

# iterate over frames continously
while True:
    # reading current frame
    successful_frame_read, frame = webcam.read()
   # boolean (always true)       #img 

    # conversion to grayscale
    grayscale_img = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    # detection algorithm
    face_detect = trained_data.detectMultiScale(grayscale_img)

    # rectangular boxes identifying faces 
    # coordinates set to x,y,w,h and color set to rgb with a border thickness of 7 
    for (x, y, w, h) in face_detect:
        cv2.rectangle(frame, (x, y), (x+w, y+h), (randrange(256), randrange(256), randrange(256)), 7)

    # displaying images + pauses the python program with the image for viewing
    cv2.imshow('Face Detector', frame)
    key = cv2.waitKey(1)

    # closes program when the q key is pressed
    if key == 81 or key == 113:
        break

print('success')