Implemented inter-frame difference method with OpenCV

Background subtraction using OpenCV

The inter-frame difference method is one of the methods for detecting moving objects. It is the result made with Python3 + OpenCV.

Click here for the original data (It's hard to understand, but it's a video that feels like splashing water on the beach)

** Click here for the binarized mask image **

** This is a video **

** Display background image **

Then it is a sample program.

movieSample.py

import cv2
import numpy as np
import time

i = 0      #Count variable
th = 30    #Difference image threshold

#Video file capture
cap = cv2.VideoCapture("/Users/.../.../.../movies/movieSample.mp4")

#Set the first frame as the background image
ret, bg = cap.read()

#Grayscale conversion
bg = cv2.cvtColor(bg, cv2.COLOR_BGR2GRAY)

while(cap.isOpened()):
    #Get frame
    ret, frame = cap.read()

    #Grayscale conversion
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    #Calculate the absolute value of the difference
    mask = cv2.absdiff(gray, bg)

    #Binarize the difference image and mask image(Monochrome)Calculate
    mask[mask < th] = 0
    mask[mask >= th] = 255

    #Display mask image
    cv2.imshow("Mask", mask)
    #Display frame image (monochrome)
    cv2.imshow("Flame", gray)
    #Display background image (monochrome)
    cv2.imshow("Background", bg)

    #Wait(0.03sec)
    time.sleep(0.03)
    i += 1    #Increase the count by 1

    #Background image update (at regular intervals)
    if(i > 30):
        ret, bg = cap.read()
        bg = cv2.cvtColor(bg, cv2.COLOR_BGR2GRAY)
        i = 0  #Initialization of count variable

    #Stops halfway when the x key is pressed
    if cv2.waitKey(1) & 0xFF == ord('x'):
        break

cap.release()
cv2.destroyAllWindows()

You may need to install OpenCV separately as a trial. Please try according to the environment. If you are using a Mac, from the terminal

$ python3 -m pip install opencv-python [ENTER] I think I can go there.

** K Igarashi ** has introduced the installation. Install OpenCV with pip