car_detector.py

# Load the Haar cascade file
import cv2

face_cascade = cv2.CascadeClassifier('cars.xml')

# Check if the cascade file has been loaded correctly
if face_cascade.empty():
    raise IOError('Unable to load the face cascade classifier xml file')

cap = cv2.VideoCapture('race2.mp4')

# Define the scaling factor
scaling_factor = 1

# Iterate until the user hits the 'Esc' key
while True:
    # Capture the current frame
    _, frame = cap.read()

    # Resize the frame
    frame = cv2.resize(frame, None,
            fx=scaling_factor, fy=scaling_factor,
            interpolation=cv2.INTER_AREA)

    # Convert to grayscale
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    # Run the car detector on the grayscale image
    face_rects = face_cascade.detectMultiScale(gray, 1.3, 5)

    # Draw a rectangle around the car
    for (x,y,w,h) in face_rects:
        cv2.rectangle(frame, (x,y), (x+w,y+h), (0,255,0), 3)

    # Display the output
    cv2.imshow('Face Detector', frame)

    # Check if the user hit the 'Esc' key
    c = cv2.waitKey(1)
    if c == 27:
        break

# Release the video capture object
cap.release()

# Close all the windows
cv2.destroyAllWindows()