\

HomeGPT: Transforming Homes with a clap, smile and a wave!!

\ This repository will follow the development of an Home Assistant project HomeGPT based on real-time embedded system. This is a post-graduate project for the University of Glasgow.

Commercial

Watch our demo on Youtube.

About

HomeGPT revolutionizes home automation with advanced features such as facial recognition for secure access, mood-responsive lighting, RFID-enabled garage entry, and clap-activated door closures. This system elevates the living experience by seamlessly blending security, convenience, and ambiance-enhancing technology.

Key Features

1. Facial Recognition based Home Entry

This feature ensures SAFETY of your home by recognizing entrants at the main door. A person can enter only if detected an ally for the people of the house. Intruders are Not Allowed as the doors won't open for you! Real Time feed of camera at the door compares an entrant's image using HOG Classifier to generate signal for welcoming the person by opening the door. You have new friend coming in, No problemo! we have feature to train the model and add new allies in real time as well!

2. Mood based Lighting Control

This state of art feature detects the mood of a person in room and adjusts the lights accordingly. Researchers believe that lighting levels have effect on the intensity of emotions. So now we can lighten up the moods and make the feel joyous in the house at all times! Camera feed detects facial gestures in real time and uses SVM Model that has been trained on Kaggle https://www.kaggle.com/datasets/jonathanoheix/face-expression-recognition-dataset dataset suing OpenCV and HOG to compare your mood and distinguish it between 7 different moods Angry, Disgust, Fear, Happy, Neutral, Sad, Surprise in real time.

3. Key less Garage Entry for Vehicles

This feature lets you sit comfortably in the car when you reach your house and lets you park the car in garage comfortably by subtarcting the hustle of getting out of the car and open the garage door*. ***RFID Sensor detects the unique identity of your car at arrival and opens the garage door for your ease of parking!

4. Clap Detection for Door Shut using FFT (Fast Fourier Transform) & DFT (Discrete Fourier Transform)

One of the stand-out feature of this project is detecting your Clap to shut that door behind you. So you can move freely in your house and ensure privacy. It's annoying if you forget to close a door behind you cause you were busy imagining your fantasy! And someone shouts from behind just to get you back into this cruel world holding accountable for manners! Real Time feed of sound is convoluted on sample sets and peaks are evaluated to judge the clap intensity. In case of match, the signal is sent to actuator to shut the door.

\

Getting Started

Clone the repository.
git clone https://github.com/savanagrawal/HomeGPT

git submodule init

git submodule update
Create build directory.
mkdir build && cd build
Build the binaries.
cmake ../

make
Run Program.
sudo ./HomeGPT --module

All the modules are listed as:

"--create-intruder-dataset" -> Create Intruder dataset
"--cid" -> Create Intruder dataset
"--train-intruder-model" -> Train Intruder model
"--tim" -> Train Intruder model
"--detect-intruder" -> Detect Intruder
"--di" -> Detect Intruder
"--detect-clap" -> Detect Claps
"--dc" -> Detect Claps
"--rfid" -> Read RFID.
"--imd" -> Start Intruder and Mood detection together.
"--d" -> All modules together.

\

Libraries / Dependencies / Softwares

Raspbian (Debian 10 : Buster/Legacy)
C++
OpenCV
CMake
Boost
PortAudio
twyleg/spidevpp - A ported version to work with C++11.
twyleg/mfrc522-cpp

\

Hardware Description

Raspberry pi 3:

Rpi

The Raspberry Pi 3 is a highly versatile and affordable single-board computer that offers impressive computing power in a compact form factor. Powered by a quad-core ARM Cortex-A53 processor running at 1.2 GHz, and equipped with 1 GB of RAM, it delivers smooth performance for a wide range of tasks.

With its various ports, including USB, HDMI, Ethernet, and GPIO, the Raspberry Pi 3 offers extensive connectivity options for peripherals and accessories, allowing for versatile integration into different projects. This makes it popular among hobbyists, educators, and DIY enthusiasts for a wide range of applications, from media centers and retro gaming consoles to robotics and smart home systems.

Overall, the Raspberry Pi 3 is a powerful and flexible single-board computer that offers a wide range of possibilities for hobbyists, students, educators, and DIY enthusiasts alike. Its compact size, affordable price, and impressive performance make it a popular choice for a wide range of computing projects and applications.

\ Raspberry pi Camera Module:

rpicam

The camera module is compact and lightweight, making it easy to integrate into various projects. It connects to the Raspberry Pi via a ribbon cable, and the camera is controlled through software using the Raspberry Pi's GPIO pins. The module features a fixed-focus lens and is capable of capturing still images with a resolution of up to 8 megapixels, as well as high-definition video with up to 1080p resolution.

The Raspberry Pi Camera Module is highly versatile and can be used for a wide range of applications, including time-lapse photography, motion detection, facial recognition, object tracking, and more. It can be used in combination with other Raspberry Pi accessories and sensors to create custom camera solutions for specific projects or applications.

\ RC522 RFID Module:

rfid

The RC522 RFID sensor is a small, low-cost module that allows for easy integration of RFID (Radio-Frequency Identification) technology with microcontroller projects. It operates at 13.56 MHz frequency and uses SPI (Serial Peripheral Interface) for communication, making it compatible with popular microcontrollers like Arduino and Raspberry Pi.

The RC522 sensor is capable of reading RFID tags and cards, and it can communicate with various types of RFID cards, including MIFARE cards, which are widely used in access control, authentication, and identification applications. It offers a reading range of a few centimeters, making it suitable for short-range applications.

\ Tower Pro Servo Motor - SG90 Digital:

servo

The Tower Pro SG90 is a digital servo motor, a type of motor that is widely used in robotics, RC (remote control) vehicles, and other applications that require precise and controlled angular movements. The SG90 is compact and lightweight, making it suitable for small-scale projects. It offers a torque of up to 1.8 kg/cm, a rotation range of 180 degrees, and operates on a voltage range of 4.8V to 6V. The digital technology in the SG90 provides improved accuracy, responsiveness, and reliability compared to analog servo motors. It is compatible with popular microcontrollers like Arduino and Raspberry Pi, and can be easily controlled using PWM (Pulse Width Modulation) signals. The Tower Pro SG90 digital servo motor is a popular choice for hobbyists and DIY enthusiasts due to its affordability, ease of use, and reliable performance in various applications.

\

Data Sheets

\

Team Members

Name	Github	GUID	Course	Role
Savan Agrawal	@savanagrawal	2781593A	MSc in Robotics and AI	Intruder, Mood and Clap detection
Chinmay Nagrale	@MiB3Avenger	2788044N	MSc in Computer Systems Engineering	Threads, Peripherals Integration and Circuit Making
Hrushikesh Sanap	@hrushisanap	2819312S	MSc in Robotics and AI	Fabrication, Documentation and Circuit Making
Mustafa Zariwala	@MustafaZari	2617991Z	MSc in Robotics and AI	Fabrication and Documentation

\

License

View our License here.

\

Code of Conduct

We follow our Code of Conduct strictly.

\

Contribution Guidelines

Want to contribute? Read our Contribution guidelines.

Contents