Abstract
As security concerns continue to rise, there is a growing demand for affordable and intelligent surveillance solutions to ensure safety in homes, businesses, and other environments. Many individuals are embracing AI-driven technologies such as Closed-Circuit Television (CCTV), smart doorbells, and automated security systems to protect their properties. This project presents a design and implementation of a cost-effective AI-powered intrusion detection system utilizing Raspberry Pi 5 for home surveillance, with adaptability for broader applications. The system integrates a camera module and an LCD screen running on a Linux-based platform, with Python, and OpenCV as key software components. It employs dlib’s deep learning-based face recognition model to detect and authenticate individuals by cross-referencing live detected faces with a stored database of known persons. Additionally, this research compares the performance of dlib-based face recognition with the YOLO model, evaluating accuracy, speed, and computational efficiency to compare the two for real-time intrusion detection.
Advisor
Dr. Rushit Dave
Committee Member
Dr. Rajeev Bukralia
Committee Member
Dr. Mansi Bhavsar
Date of Degree
2025
Language
english
Document Type
APP
Degree
Master of Science (MS)
Program of Study
Data Science
Department
Computer Information Science
College
Science, Engineering and Technology
Recommended Citation
Osiemo, M. N. (2025). Design and implementation of a low-cost Raspberry Pi and AI-based intrusion detection system for surveillance [Master’s alternative plan paper, Minnesota State University, Mankato]. Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University, Mankato. https://cornerstone.lib.mnsu.edu/etds/1484/
Rights Statement
