Webcam Motion Detector Program in Python

In this we are going to write python program which is going to analyse the images taken from the webcam and try to detect the movement and store the time-interval of the webcam video in a csv file.

Required Library

We are going to use the OpenCV & pandas library for that. If it’s not already installed, you can install it using pip, with something like:

$pip install opencv2, pandas

Example Code

#Import required libraries
import cv2
import pandas as pd
import time
from datetime import datetime

#Initialise variables
stillImage = None
motionImage = [ None, None ]
time = []

# Initializing the DataFrame with start and end time
df = pd.DataFrame(columns = ["start", "end"])

# Capturing video
video = cv2.VideoCapture(0)

while True:
   # Start reading image from video
   check, frame = video.read()
   motion = 0

   # Convert color image to gray_scale image
   gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
   gray = cv2.GaussianBlur(gray, (21, 21), 0)
   if stillImage is None:
      stillImage = gray
      continue
   # Still Image and current image.
   diff_frame = cv2.absdiff(stillImage, gray)

   # change the image to white if static background and current frame is greater than 25.
   thresh_frame = cv2.threshold(diff_frame, 25, 255, cv2.THRESH_BINARY)[1]
   thresh_frame = cv2.dilate(thresh_frame, None, iterations = 2)
   # Finding contour and hierarchy from a moving object.
   contours,hierachy = cv2.findContours(thresh_frame.copy(),
      cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
   for contour in contours:
      if cv2.contourArea(contour) < 10000:
         continue
      motion = 1
      (x, y, w, h) = cv2.boundingRect(contour)
      cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 3)
   # Append current status of motion
   motionImage.append(motion)
   motionImage = motionImage[-2:]
   # Append Start time of motion
   if motionImage[-1] == 1 and motionImage[-2] == 0:
      time.append(datetime.now())

   # Append End time of motion
   if motionImage[-1] == 0 and motionImage[-2] == 1:
      time.append(datetime.now())
   # Displaying image in gray_scale
   cv2.imshow("Gray_Frame", gray)

   # Display black and white frame & if the intensity difference is > 25, it turns white
   cv2.imshow("Threshold Frame", thresh_frame)
   # Display colored frame
   cv2.imshow("Colored_Frame", frame)

   key = cv2.waitKey(1)
   # Press q to stop the process
   if key == ord('q'):
      if motion == 1:
         time.append(datetime.now())
      break

# Append time of motion
for i in range(0, len(time), 2):
   df = df.append({"Start":time[i], "End":time[i + 1]}, ignore_index = True)

# Creating a csv file in which time of movements will be saved
df.to_csv("FrameInMotion_time.csv")

video.release()

# close window
cv2.destroyAllWindows()

Output

We can see we’ll get 3 different Windows which are going to display our current movement from webcam in three different modes (grayscale, colored & black&White).

It will also store the datetime of our webcam Motion in a csv and our output from csv will be something like:

FrameMotion_time.csv (output)

start     end      End                            Start
0                  2019-02-21 18:10:59.718005     2019-02-21 18:08:35.791487