Face detection using Haar Cascades – OpenCV 3.4 with python 3 Tutorial 37

You will learn in this video how to detect Faces using the Haar Cascades object detection method.

Source code:

import cv2
import numpy as np
def nothing(x):
    pass
cap = cv2.VideoCapture(0)
face_cascade = cv2.CascadeClassifier("haarcascade_frontalface_default.xml")
cv2.namedWindow("Frame")
cv2.createTrackbar("Neighbours", "Frame", 5, 20, nothing)
while True:
    _, frame = cap.read()
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    neighbours = cv2.getTrackbarPos("Neighbours", "Frame")
    faces = face_cascade.detectMultiScale(gray, 1.3, neighbours)
    for rect in faces:
        (x, y, w, h) = rect
        frame = cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)

    cv2.imshow("Frame", frame)
    key = cv2.waitKey(1)
    if key == 27:
        break
cap.release()
cv2.destroyAllWindows()

Files:

1. haarcascades.zip