NOTE: Used a pre-trained model
Face detection in photos can be performed using the classical feature-based cascade classifier using the OpenCV library (cv2).
Used packages
import cv2
import time
import os
OpenCV provides the CascadeClassifier class that creates a cascade classifier for face detection. The constructor can take a filename as an argument that specifies the XML file for a pre-trained model.
Download a pre-trained model (file here) for frontal face detection from the OpenCV GitHub project – to be placed in your working directory.
Setup the image directory
# function to get images from folder
def get_images(dir_name):
list_images = os.listdir(dir_name)
all_images =list()
for entry in list_images:
full_path =os.path.join(dir_name, entry)
if os.path.isdir(full_path):
all_images.all_images + get_images(full_path)
else: all_images.append(full_path)
return all_images
Load the model that will perform face detection on photographs by calling the detectMultiScale() function.
# Face Detection
def main():
dir_name = 'images' # directory for images
list_images = get_images(dir_name)
for i in range(20): #20 images
image_path = list_images[i]
print(image_path)
# load the pre-trained model
case_path = "haarcascade_frontalface_default.xml" # define the model used for recognition detection
faceCascade= cv2.CascadeClassifier(case_path)
image=cv2.imread(image_path)
gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY) # make the picture to gray from color
# face detection
faces = faceCascade.detectMultiScale(gray,
scaleFactor = 1.1,
minNeighbors=5,
minSize = (30, 30))
for (x, y, w, h) in faces: # draw rectangles on the faces when detected
cv2.rectangle(image, (x,y), (x + w, y+h), (0, 255, 0), 2)
#Load the detected faces
cv2.imshow("Face Found", image)
cv2.waitKey(5)
time.sleep(5)
cv2.destroyAllWindows()
if __name__ == '__main__':
main()
Running the model loads the images and configures the cascade classifier; faces are detected, and each bounding box gets printed.
The model only works for faces directly at the camera (in front).
