Turbo Search

Inspiration

I was annoyed by having to search for quiz questions that were in the form of an image, and being forced to type out the text

What it does

Turbo Search allows the user to take a screenshot of an image with text in it and extracts the text. That text can then be searched on Google, copied to the clipboard, or saved to a file

How we built it

We used Java for the entire project, including the screenshots, character recognition, and GUI. We also used a Java-wrapped Optical Character Recognition to extract the text

Challenges we ran into

A major setback came when incorporating the trained Optical Character Recognition database

Accomplishments that we're proud of

Creating a fully functioning application within 24 hours

What we learned

How to implement OCR for text extraction, as well as working on smaller parts of a program together as a team

What's next for Turbo Search

Designing and creating a Chrome and Firefox plugin to extend functionality to the web

Built With

Submitted to

FloridaTechHacks

Created by

I developed the screencapture view and output menu. I also brought all of the code together to form a cohesive and functional project

Dmitri Piquero
I began implementing the screenshot feature that overlays the user screen and allows the user to take a snippet section using a mouse. I also played a role in the implementation of the java OCR that was used in the final project.

Joshua Jennings
I provided the project idea and worked on how it will flow together. I also provided a prototype of the optical character recognition system to see how we need to pivot our model.

Jonathan Wamsley
Collaborated on researching and implementing the OCR functionality for the project. Researched potential to utilize either python or matlab as an alternative to JAVA

Rick Ammon