Motion detection with camera in Raspberry Pi

Introduction

Motion detection with a camera helps you know when something moves in front of the camera. It is useful for security or monitoring.

To watch your home and get alerts if someone moves inside.

To monitor a pet playing in a room and record only when it moves.

To save storage by recording video only when there is motion.

To count people entering or leaving a room.

To trigger lights or alarms when motion is detected.

Syntax

Raspberry Pi

import cv2

# Start camera
camera = cv2.VideoCapture(0)

# Read first frame
ret, frame1 = camera.read()
frame1_gray = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY)
frame1_gray = cv2.GaussianBlur(frame1_gray, (21, 21), 0)

while True:
    ret, frame2 = camera.read()
    frame2_gray = cv2.cvtColor(frame2, cv2.COLOR_BGR2GRAY)
    frame2_gray = cv2.GaussianBlur(frame2_gray, (21, 21), 0)

    # Calculate difference
    diff = cv2.absdiff(frame1_gray, frame2_gray)
    _, thresh = cv2.threshold(diff, 25, 255, cv2.THRESH_BINARY)

    # Find contours
    contours, _ = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

    for contour in contours:
        if cv2.contourArea(contour) > 500:
            print("Motion detected!")

    frame1_gray = frame2_gray

    if cv2.waitKey(10) & 0xFF == 27:  # Press ESC to exit
        break

camera.release()
cv2.destroyAllWindows()

This example uses OpenCV library to access the camera and process images.

We compare two frames to find changes that show motion.

Examples

Start the camera and prepare the first image in gray and blurred to reduce noise.

Raspberry Pi

import cv2

camera = cv2.VideoCapture(0)
ret, frame1 = camera.read()
frame1_gray = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY)
frame1_gray = cv2.GaussianBlur(frame1_gray, (21, 21), 0)

Find the difference between two frames and create a black and white image showing changes.

Raspberry Pi

diff = cv2.absdiff(frame1_gray, frame2_gray)
_, thresh = cv2.threshold(diff, 25, 255, cv2.THRESH_BINARY)

Look for shapes in the difference image that are big enough to count as motion.

Raspberry Pi

contours, _ = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
for contour in contours:
    if cv2.contourArea(contour) > 500:
        print("Motion detected!")

Sample Program

This program uses your Raspberry Pi camera to detect motion. It compares each new frame to the previous one. If it finds big enough changes, it prints "Motion detected!". Press ESC to stop.

Raspberry Pi

import cv2

camera = cv2.VideoCapture(0)
ret, frame1 = camera.read()
frame1_gray = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY)
frame1_gray = cv2.GaussianBlur(frame1_gray, (21, 21), 0)

while True:
    ret, frame2 = camera.read()
    frame2_gray = cv2.cvtColor(frame2, cv2.COLOR_BGR2GRAY)
    frame2_gray = cv2.GaussianBlur(frame2_gray, (21, 21), 0)

    diff = cv2.absdiff(frame1_gray, frame2_gray)
    _, thresh = cv2.threshold(diff, 25, 255, cv2.THRESH_BINARY)

    contours, _ = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

    motion_found = False
    for contour in contours:
        if cv2.contourArea(contour) > 500:
            motion_found = True
            break

    if motion_found:
        print("Motion detected!")

    frame1_gray = frame2_gray

    if cv2.waitKey(10) & 0xFF == 27:  # ESC key
        break

camera.release()
cv2.destroyAllWindows()

Important Notes

Make sure your Raspberry Pi camera is enabled and connected.

You need to install OpenCV with: pip install opencv-python.

Adjust the contour area threshold (500) to make detection more or less sensitive.

Summary

Motion detection compares camera images to find changes.

OpenCV helps capture and process images easily.

Detecting motion can trigger alerts or recordings.