Inspiration

I was inspired to create VIDText by looking at my own experiences in the education system. Many of the homework and study assignments are provided through a hardcopy form. Moving this to a digital frame is tedious and can be easily streamlined by text extraction solutions like VIDText.

I believe that text extraction provides a better way to save content online, allow for easier work (text can be copied and pasted in places such as Google Docs) and makes manipulation possible. These are all advantages that text extraction has over a simple image.

What it does

VIDText extracts text from an image and pastes it into the command terminal. The application is seamlessly integrated into a simple and interactive GUI (Graphical User Interface) to allow for quick and easy access.

How we built it

This application was built using computer vision techniques and image processing functions. I used certain external libraries such as opencv and pytesseract. Additonally I used tkinter to create the GUI. This entire program was written in Python in the given time period.

Challenges we ran into

I am proud of overcoming certain challenges, such as when the main thread was being overloaded by the video feed, disabling user interaction with the GUI momentarily. I solved this by moving the video processes on a seperate thread, significantly boosting performance and user experience.

Accomplishments that we're proud of

I am proud of being able to successfully integrate text extraction into a user-friendly GUI application. I think that this tool, with a few improvements, could be very useful for students and teachers alike in the educational system!

What we learned

I learned a lot about computer vision techniques such as OCR (Optical Character Recognition). I was able to match text in an image to values in an internal database with pattern matching algorithms.

What's next for VIDText

VIDText can definetly be improved! I will work on text extraction accuracy and gui improvements. New features can also boost user experience!

Built With

Share this project:

Updates