Motion detection with camera in Raspberry Pi - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this motion detection frame difference code?

Consider this Python snippet using OpenCV on Raspberry Pi to detect motion by frame difference. What will be printed when no motion is detected?

Raspberry Pi

import cv2

frame1 = cv2.imread('frame1.jpg', 0)
frame2 = cv2.imread('frame2.jpg', 0)
diff = cv2.absdiff(frame1, frame2)
non_zero_count = cv2.countNonZero(diff)

if non_zero_count > 1000:
    print('Motion detected')
else:
    print('No motion')

A No motion

B Motion detected

C cv2.error: OpenCV(…) error

D TypeError: unsupported operand type(s)

💡 Hint

Think about what happens when the two frames are identical or very similar.

If the two frames are almost the same, the difference image will have very few non-zero pixels, so the count will be less than 1000, printing 'No motion'.

🧠 Conceptual

intermediate

Which sensor setup is best for detecting motion with a Raspberry Pi camera?

You want to detect motion using a Raspberry Pi camera module. Which setup will give the most reliable motion detection?

A Use a temperature sensor to detect heat changes

B Compare consecutive frames and check pixel differences

C Use a single frame and check if it is completely black

D Capture one frame per hour and compare with a stored image

💡 Hint

Motion means change over time in the camera view.

Motion detection works best by comparing consecutive frames to find differences. A single frame or very slow captures won't detect motion well.

🔧 Debug

advanced

Why does this Raspberry Pi motion detection code raise an error?

This code snippet is intended to detect motion but raises an error. What is the cause?

Raspberry Pi

import cv2

cap = cv2.VideoCapture(0)
ret, frame1 = cap.read()
ret, frame2 = cap.read()
diff = cv2.absdiff(frame1, frame2)
gray = cv2.cvtColor(diff, cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (5,5), 0)
_, thresh = cv2.threshold(blur, 20, 255, cv2.THRESH_BINARY)
contours, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)

for contour in contours:
    if cv2.contourArea(contour) < 1000:
        continue
    cv2.drawContours(frame1, [contour], -1, (0,255,0), 2)

cv2.imshow('Motion', frame1)
cv2.waitKey(0)
cv2.destroyAllWindows()

A cv2.VideoCapture(0) failed to open the camera

B cv2.drawContours requires grayscale image, not color

C cv2.findContours returns 3 values, not 2

D cv2.threshold returns only one value, not two

💡 Hint

Check the number of values returned by cv2.findContours in your OpenCV version.

In OpenCV 3.x, cv2.findContours returns 3 values: image, contours, hierarchy. The code expects 2 values, causing a ValueError.

📝 Syntax

advanced

Which option fixes the syntax error in this Raspberry Pi motion detection snippet?

Fix the syntax error in this dictionary comprehension used to count motion pixels: motion_pixels = {x: x*2 if x > 5 for x in range(10)}

A motion_pixels = {x: (x*2 if x > 5 else 0) for x in range(10)}

B motion_pixels = {x: x*2 for x in range(10)}

C motion_pixels = {x: x*2 if x > 5 for x in range(10)}

D motion_pixels = {x: x*2 for x in range(10) if x > 5}

💡 Hint

Remember the correct order of if conditions in dictionary comprehensions.

Option D correctly places the if condition after the for loop to filter keys. Option D uses a ternary but includes else, which changes the output.

🚀 Application

expert

How many contours will be detected by this Raspberry Pi motion detection code?

Given this code snippet, how many contours will be detected if the camera captures a static scene with one moving object covering a large area?

Raspberry Pi

import cv2

# Assume frame1 and frame2 are consecutive frames
# with one large moving object

frame1 = cv2.imread('static_scene.jpg')
frame2 = cv2.imread('moving_object.jpg')
diff = cv2.absdiff(frame1, frame2)
gray = cv2.cvtColor(diff, cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (5,5), 0)
_, thresh = cv2.threshold(blur, 25, 255, cv2.THRESH_BINARY)
contours, _ = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

large_contours = [c for c in contours if cv2.contourArea(c) > 5000]
print(len(large_contours))

A 1

B 0

C Multiple small contours

D Raises an error

💡 Hint

Think about how a large moving object appears in the thresholded image.

A large moving object creates one big contour above the area threshold, so the list will contain one contour.