By leveraging the power of existing libraries for computer vision, it is possible to write (apparently) quite simple scripts to automate menial image processing tasks. Following is how we can write a Python script for detecting faces in image files with the OpenCV python wrapper.
Installing OpenCV
I choose to install OpenCV with pip. Doing it this way is as simple as running the following in your terminal.
pip install --user opencv-python
The --user switch is for performing a local install for current user account.
Detecting Faces
Following is a Python script written for the task.
#!/usr/bin/env python3
"""
Detect Faces in a given image and save new image with faces marked.
"""
import cv2
import sys
def detect_faces(image_path):
# Locate haarcascade file
casc_path = "haarcascade_frontalface_default.xml"
# Create classifier based on located haarcascade
face_classifier = cv2.CascadeClassifier(casc_path)
image = cv2.imread(image_path)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# Detect faces in image
faces = face_classifier.detectMultiScale(
gray,
scaleFactor=1.1,
minNeighbors=5,
minSize=(150, 150)
)
print("Found {0} faces".format(len(faces)))
# Draw rectangles around the faces
for (x, y, w, h) in faces:
cv2.rectangle(image, (x, y), (x+w, y+h), (0, 255, 0), 2)
cv2.imshow("Faces found", image)
status = cv2.imwrite("{0}_detected.{1}".format(image_path.split(".")[0], image_path.split(".")[1]), image)
print ("Image written to file-system: ", status)
cv2.waitKey(0)
def main():
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("-i", "--img",
help="Path to input image")
detect_faces((parser.parse_args()).img)
if __name__ == "__main__":
main()Give the script (i.e. detect_faces_img.py) permission to execute with:
chmod u+x face_detection_img.py
Run the script from residing directory as ./face_detection_img.py image.jpg to detect faces in image.jpg.