My PBL Project

Investigation Questions:

How can machine learning algorithms be used to improve the accuracy and efficiency of a security camera system?

Machine learning algorithms can be used to improve the accuracy and efficiency of a security camera system in several ways. One way is by using object detection algorithms, which can identify and classify objects within the camera's field of view. This can be used to trigger alerts or take specific actions when certain objects, such as a person or vehicle, are detected. Machine learning algorithms can also be used to analyze patterns and trends in the data collected by the security camera system, allowing it to make more informed decisions and respond more effectively to potential threats. Additionally, machine learning can be used to optimize the performance of the security camera system by adapting to changing conditions and improving over time.

How can we ensure that the security camera system respects the privacy of individuals being monitored?

Limit the scope of surveillance: Only monitor areas where there is a legitimate need for security, and avoid unnecessarily collecting data on individuals.

Obtain consent: Where possible, obtain the consent of individuals before collecting their data. This may involve providing clear notice and obtaining explicit consent for the collection and use of their data.

Use de-identification techniques: Where it is not possible to obtain consent, consider using de-identification techniques to protect the privacy of individuals. These techniques can include blurring or obscuring faces or other identifying features, or using machine learning algorithms to automatically remove such information from the data.

What are the most cost-effective hardware and software options for implementing a security camera system with OpenCV?

The most cost-effective hardware and software options for implementing a security camera system with OpenCV will depend on the specific requirements and goals of your system. Some factors to consider when selecting hardware and software include the resolution and frame rate of the cameras, the processing power required for the desired machine learning algorithms, and the type of storage and networking infrastructure needed to store and transmit the data.

How will our security camera identify faces?

We will use the haar cascade classifier in this project. Haar-cascade classifier is an Object Detection Algorithm used to identify faces in an image or a real-time video. The algorithm uses edge or line detection features proposed by Viola and Jones in their research paper “Rapid Object Detection using a Boosted Cascade of Simple Features” published in 2001. The algorithm is given a lot of positive images consisting of faces, and a lot of negative images not consisting of any face to train on them. The model created from this training is available at the OpenCV.

Why use the coding language "Python" instead of other languages?

We used the coding language Python because it is free and accessible to everyone. In addition, we used Python because it has a variety of features such as lots of libraries. Lastly, Python is very simple and easy to use.

Our Code:

import cv2

import time

import datetime

import os

from email.message import EmailMessage

import ssl

import smtplib

import imghdr

email_sender = 'preston.hsadvancement22@gmail.com'

email_password = "User's Password"

email_receiver = 'preston.hsadvancement22@gmail.com'

subject = 'Face Detected'

body = """

Face Detected

"""

em = EmailMessage()

em['From'] = email_sender

em['To'] = email_receiver

em['Subject'] = subject

em.set_content(body)

cap = cv2.VideoCapture(0)

face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_frontalface_default.xml")

body_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascadefullbody.xml")

frame_size = (int(cap.get(3)), int(cap.get(4)))

detection = False

detection_stopped_time = None

timer_started = False

seconds_after_detection = 5

while True:

_, frame = cap.read()

gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

faces = face_cascade.detectMultiScale(gray, 1.3, 5)

bodies = face_cascade.detectMultiScale(gray, 1.3, 5)

if len(faces) + len(bodies) > 0:

if detection:

timer_started = False

else:

detection = True

current_time = datetime.datetime.now().strftime("%d-%m-%Y-%H-%M-%S")

imagefile = f"{current_time}.jpg"

s, img = cap.read()

cv2.imwrite(imagefile, img) # save image

print("Took Picture")

elif detection:

if timer_started:

if time.time() - detection_stopped_time > seconds_after_detection:

detection = False

timer_started = False

print("stop recording")

with open(imagefile, 'rb') as f:

file_data = f.read()

file_type = imghdr.what(f.name)

file_name = f.name

em.add_attachment(file_data, maintype='png', subtype=file_type, filename=file_name)

context = ssl.create_default_context()

with smtplib.SMTP_SSL('smtp.gmail.com', 465, context=context) as smtp:

smtp.login(email_sender, email_password)

smtp.sendmail(email_sender, email_receiver, em.as_string())

print ("sent email")

else:

timer_started = True

detection_stopped_time = time.time()

for (x, y, width, height) in faces:

cv2.rectangle(frame, (x, y), (x + width, y + height), (255, 0, 0), 3)

cv2.imshow("camera", frame)

if cv2.waitKey(1) == ord("q"):

break

cap.release()

cv2.destroyAllWindows()