You can do the following without installation

Input image (original image)

The output image (image after conversion) and the position of the cat's face are recognized, and a circle is added there.

You will need a Google account as you will be using Google Colaboratory.

1. Connect to Google Colaboratory

Search for "Google Drive" and sign in with your Google account.

Select "New", "Other", and "Add App".

Select Colaboratory.

Click Continue to install Colaboratory.

After installation, you can select "Google Colaboratory" from "New" and "Other".

If you select "Google Colaboratory", a new file will be created and Colaboratory will start.

From this point onward, everything except preparing the necessary files will be performed within the Colaboratory.

2. OpenCV, loading drawing library (matplotlib)

Type the following in the first cell of the Colaboratory and press Shift + Enter. It is OK if the OpenCV version (4.1.2 as of 2020.08.17) is displayed.

%matplotlib inline #Inline matplotlib output(Not a pop-up)To.
import cv2 #load opencv
import matplotlib.pyplot as plt #Load matplotlib and use pyplot with the name plt.
print(cv2.__version__) #Display the version of opencv and check the loading.
#output
# 4.1.2

3. Prepare the necessary files

Download the trained data from https://github.com/opencv/opencv/tree/master/data/haarcascades ..

For now, click on the haarcascade_frontalcatface.xml file to recognize the cat's face.

Right-click on Raw to save the file.

4. Upload the file to Colaboratory

Write the following code in the second cell and press Shift + Enter.

from google.colab import files #Use file operations from the colaboratory library.
f = files.upload() #Upload the file.

Press "Select File" to upload the file.

The file uploads the following two.

Image file you want to convert
haarcascade_frontalcatface.xml (Prepare the necessary files, the file downloaded in)

Write the following code in the next cell, press Shift + Enter and execute it, and check that the name of the uploaded file is output (in this example, the file name of the image you want to convert is "IMG_20200808_181512" .jpg ").

!ls #Execute the system command ls to check if the file is uploaded.
#output
# haarcascade_frontalcatface.xml	IMG_20200808_181512.jpg  sample_data

5. Load images with OpenCV

Write the following code in a new cell and press Shift + Enter to execute it.

#file name(IMG_20200808_181512.jpg)Please change to the file name of the image you uploaded.
img = cv2.imread("./IMG_20200808_181512.jpg ") #Loading images
show_img = cv2.cvtColor(img,cv2.COLOR_BGR2RGB) #Create an image with BGR format converted to RGB format for display.
plt.imshow(show_img) # matplot.Display the image read using lib.

It is OK if the uploaded image is displayed as shown below.

6. Face recognition and draw a circle and an arrow at the position of the face

Write the following code in a new cell and press Shift + Enter to run it. For details on cv2.circle and cv2.putText, see [OpenCV drawing function](http://labs.eecs.tottori-u.ac.jp/sd/Member/oyamada/OpenCV/html/py_tutorials/py_gui/py_drawing_functions/ Please refer to py_drawing_functions.html) etc.

#Generate a model from a trained file
face_cascade = cv2.CascadeClassifier('./haarcascade_frontalcatface.xml')
#Generate gray image
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

#Face recognition
faces = face_cascade.detectMultiScale(gray, 1.3, 5)

#Loop with recognition result x,y:Upper left of face rectangle, w,h:Width and height of face rectangle
for (x,y,w,h) in faces:
  #Draw a circle at the position of the face
  img = cv2.circle(img,(x+int(w/2),y+int(h/2)),int(max(w/2,h/2)),(0,0,255),12)
  #Draw an arrow with letters
  cv2.putText(img,"<---",(x+w,y+int(h/2)),cv2.FONT_HERSHEY_SIMPLEX,8,(0,0,255),32,cv2.LINE_AA)

#Convert to RGB for drawing
show_img = cv2.cvtColor(img,cv2.COLOR_BGR2RGB)
#Drawing the result
plt.imshow(show_img)

If the cat's face is circled and has an arrow as shown below, it is successful. However, there are times when the face cannot be recognized. For example, in this image, the calico cat is facing you behind, but the face is not recognizable.

7. Other

Please refer to sites such as OpenCV-Python Tutorial.

[PYTHON] Try OpenCV with Google Colaboratory