Esp32cam alert using python

import urllib.request
import cv2 as cv
import numpy as np
from twilio.rest import Client

# Set up Twilio account credentials
account_sid = 'Enter your twilio id'
auth_token = 'Enter your twilio token'
client = Client(account_sid, auth_token)
twilio_number = 'Enter your twilio phone number'
destination_number = 'Enter your phone number'

# IP camera URL
url = 'http://192.168.5.110/capture?' ##enter your esp32cam ip address

# Initialize variables
frame1 = None
motion_detected = False

print('START')

while True:
    # Read image from IP camera using HTTP
    imgResponse = urllib.request.urlopen(url)
    imgNp = np.array(bytearray(imgResponse.read()), dtype=np.uint8)
    frame2 = cv.imdecode(imgNp, -1)

    # If there's an error reading from camera, break loop
    if frame2 is None:
        break

    # Convert to grayscale and blur
    gray = cv.cvtColor(frame2, cv.COLOR_BGR2GRAY)
    gray = cv.GaussianBlur(gray, (21, 21), 0)

    # If there's no previous frame, set current frame as previous frame and continue loop
    if frame1 is None:
        frame1 = gray
        continue

    # Compute absolute difference between current and previous frame
    frame_delta = cv.absdiff(frame1, gray)

    # Apply thresholding to eliminate noise and convert to binary image
    thresh = cv.threshold(frame_delta, 25, 255, cv.THRESH_BINARY)[1]
    thresh = cv.dilate(thresh, None, iterations=2)

    # Find contours in the binary image
    contours, _ = cv.findContours(thresh.copy(), cv.RETR_EXTERNAL, cv.CHAIN_APPROX_SIMPLE)

    # Loop over contours
    for contour in contours:
        # If contour area is too small, ignore it
        if cv.contourArea(contour) < 1000:
            continue

        # Draw a bounding box around the contour and mark motion detected
        (x, y, w, h) = cv.boundingRect(contour)
        cv.rectangle(frame2, (x, y), (x + w, y + h), (0, 0, 255), 2)
        motion_detected = True

    # Display current frame and motion detection status
    cv.imshow("IP Camera", frame2)
    cv.imshow("Thresh", thresh)
    key = cv.waitKey(1) & 0xFF

    # If 'q' is pressed, break loop
    if key == ord("q"):
        break

    # If motion detected, send an SMS alert
    if motion_detected:
        print("Motion detected!")
        message = client.messages.create(
            body='Motion detected!',
            from_=twilio_number,
            to=destination_number
        )
        print(message.sid)
        motion_detected = False

    # Set current frame as previous frame for next iteration
    frame1 = gray

# Release VideoCapture object and close windows
cv.destroyAllWindows()
print('END')