While trying to make an outside pet house for the pet, we tried to make the pet house smart so that the pet detection system grants access only when it's picture matches the database.

What it does

It's a pet detection system, which uses led lights, ultra-sonic sound sensors, web camera, a step-up motor connected to an Arduino Uno board. The ultrasonic sensor measures the distance of the pet from the door and blinks green when an object is at a certain distance away. The green light triggers the webcam which uses machine learning algorithms to check if the object's image exists in the database. If so, the step-up motor rotates to open the door allowing access to the pet.

How we built it

We used Arduino Uno, Step-up motor, breadboard, Ultrasonic sensor, pulley, Webcam, a wooden framework for the hardware. We used Python, Tensorflow, openCV, Tkinter, Kates for the GUI and other Software applications. We used HTML5, CSS3, jQuery, JavaScript, Node.js, Express, Bootstrap framework for making the website.

Challenges we ran into

One of the problems we ran into was triggering the webcam on to take a picture whenever the object is at a certain distance.

Accomplishments that we're proud of

In this project we were able to accomplish a lot, we were able to create the entire hardware from scratch and use web development and machine learning! As sophomores, we take immense pride in order to take on such a project!

What we learned

We learned a lot of new things in doing this project. We learned how to make the neural network learn better and predict with higher efficiency make a parallax website and how to make a GUI application using Tkinter which has the control of all the applications we are doing.

What's next for Project Portal

We plan to work with this project even after this competition and would like to improvise on what we already did.

Built With

Share this project: