Face Recognition for Beginner (Part 1) Using Python

Know how a computer captures a video image in the form of a pixel value matrix

On this occasion, we will try to do Face Recognition using python from the basics of knowing how computers capture images.

First, we import the necessary modules (make sure you have previously installed opencv-python). I write the code using Notepad ++.

# Import module 
import cv2, time

then make a command to capture the object from our computer camera then close it

video = cv2.VideoCapture(0)
video.release()

thus, the whole command looks like this,

# Import module 
import cv2, timevideo = cv2.VideoCapture(0)
video.release()

then save the code with a name, for example, “capture1.py” in a folder, for example, C:Usersasuscapture1

then open a command prompt and run the following command,

C:Usersasuscapture1>python capture1.py

if there is no error or error message from the command, then we can continue.

Next, we create an object frame to capture the image from the camera

import cv2, time
camera = 0
video = cv2.VideoCapture(camera, cv2.CAP_DSHOW)
check, frame = video.read()
print(check)
print(frame)
video.release()

save the file with the same name before which is capture1.py,

then open a command prompt and run the following command,

C:Usersasuscapture1>python capture1.py

you will see the following result

Then you can see the screenshot of the pixel values ​​in the matrix.

Then we add a command to capture the image frame once and add a command to exit the image capture

import cv2, time
camera = 0
video = cv2.VideoCapture(camera,cv2.CAP_DSHOW)
check, frame = video.read()
print(check)
print(frame)
cv2.imshow("capture1", frame)
cv2.waitKey(0)
video.release()

save it again with the same name, then open a command prompt and run the following command (make sure the camera on your computer is in good condition and good lighting. If you are in dark lighting then the image capture will not appear, the alternative is to light your face with a flashlight)

C:Usersasuscapture1>python capture1.py

then an image will appear as follows

You can see the captured camera and the pixel value matrix, then we convert the image to gray,

import cv2, timecamera = 0
video = cv2.VideoCapture(camera, cv2.CAP_DSHOW)check, frame = video.read()print(check)
print(frame)gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)cv2.imshow("capture1", gray)
cv2.waitKey(0)video.release()

then an image will appear as follows

# Import module 
import cv2, time# capture command with camera
camera = 0
video = cv2.VideoCapture(camera, cv2.CAP_DSHOW)
a=0
while True :
a = a+1
check, frame = video.read()print(check)
print(frame)gray=cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)cv2.imshow("capture1",gray)
#cv2.waitKey(0)key = cv2.waitKey(1)
if key == ord('q') :
breakprint(a)
# Command closes camera
video.release()cv2.destroyAllWindows()