Face Detection on Videos using OpenCV Haar Cascades

This post will be helpful in learning OpenCV using Python programming. Here I will show how to implement OpenCV functions and apply it in various aspects using some examples. Then the output will be shown with some comparisons as well.

Requirements:

OpenCV 3.4
Python 3.6+
Numpy
Image, Webcam or Video input
Documentation Source: OpenCV Official Documentation

First, you need to setup your Python Environment with OpenCV. You can easily do it by following Life2Coding’s tutorial on YouTube: Linking OpenCV 3 with Python 3

Goals:

In this tutorial, I will show you how to detect the faces on webcam videos or normal videos using OpenCV Haar Cascades and Python coding.

Documentation:

Python: cv2.CascadeClassifier.detectMultiScale(image[, scaleFactor[, minNeighbors[, flags[, minSize[, maxSize]]]]])→ objects

Detects objects of different sizes in the input image. The detected objects are returned as a list of rectangles

Parameters:

cascade – Haar classifier cascade (OpenCV 1.x API only). It can be loaded from XML or YAML file using Load(). When the cascade is not needed anymore, release it using cvReleaseHaarClassifierCascade(&cascade).
image – Matrix of the type CV_8U containing an image where objects are detected.
objects – Vector of rectangles where each rectangle contains the detected object, the rectangles may be partially outside the original image.
numDetections – Vector of detection numbers for the corresponding objects. An object’s number of detections is the number of neighboring positively classified rectangles that were joined together to form the object.
scaleFactor – Parameter specifying how much the image size is reduced at each image scale.
minNeighbors – Parameter specifying how many neighbors each candidate rectangle should have to retain it.
flags – Parameter with the same meaning for an old cascade as in the function cvHaarDetectObjects. It is not used for a new cascade.
minSize – Minimum possible object size. Objects smaller than that are ignored.
maxSize – Maximum possible object size. Objects larger than that are ignored.

Steps:

Initialize webcam feed using cv2.VideoCapture()
Read webcam images using cv2.VideoCapture. read()

Download the Haar Cascade: https://github.com/opencv/opencv/blob/master/data/haarcascades/haarcascade_frontalface_alt.xml

Load the Face Haar Cascade file using cv2.CascadeClassifier()
Detects the face on the image using cv2.CascadeClassifier.detectMultiScale()
Draw boundary around the face using cv2.rectangle()
Display each frames of the video using cv2.imshow()
Wait for keyboard button press using cv2.waitKey()
Release the VideoCapture using cv2.VideoCapture.release()
Exit window and destroy all windows using cv2.destroyAllWindows()

Example Code:

import cv2

scale_factor = 1.2
min_neighbors = 3
min_size = (50, 50)
webcam=True #if working with video file then make it 'False'

def detect(path):

    cascade = cv2.CascadeClassifier(path)
    if webcam:
        video_cap = cv2.VideoCapture(1) # use 0,1,2..depanding on your webcam
    else:
        video_cap = cv2.VideoCapture("videoFile.mp4")
    while True:
        # Capture frame-by-frame
        ret, img = video_cap.read()

        #converting to gray image for faster video processing
        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

        rects = cascade.detectMultiScale(gray, scaleFactor=scale_factor, minNeighbors=min_neighbors,
                                         minSize=min_size)
        # if at least 1 face detected
        if len(rects) >= 0:
            # Draw a rectangle around the faces
            for (x, y, w, h) in rects:
                cv2.rectangle(img, (x, y), (x + w, y + h), (0, 255, 0), 2)

            # Display the resulting frame
            cv2.imshow('Face Detection on Video', img)
            #wait for 'c' to close the application
            if cv2.waitKey(1) & 0xFF == ord('c'):
                break
    video_cap.release()

def main():
    cascadeFilePath="haarcascade_frontalface_alt.xml"
    detect(cascadeFilePath)
    cv2.destroyAllWindows()


if __name__ == "__main__":
    main()