Hangman OCR

• 

  trying out possible combinations

• 

  reading in my chicken scratch hand writing

Inspiration

We were inspired by the growing wave of AI being used at such an everyday level and wanted to show how this innovation lowers the barrier even more for human computer interaction.

What it does

It takes a picture of written letters in an incomplete hang man game and parses the data into text using optical character recognition. It then provides a list of possible solutions to the puzzle.

How we built it

We used React for the front end framework and java for the logic handling on the backend.

Challenges we ran into

We ran into trouble getting the AI to distinguish between what it was supposed to be reading and what was just background noise.

Accomplishments that we're proud of

We are proud of the sturdiness of the final website and the awesome optical abilities that it has. We hope to expand on this greatly next year.

What we learned

We learned how to implement the Tesseract.js framework and also access the user's camera, bypassing SSR. We also learned how to permutate through combinations of possible strings to come up with a list of probable answers to the game.

What's next for Hangman OCR

In the future we are planning to beef up its OCR distinguishing skills so it can more seamlessly take in the information we need it to. We want to expand this OCR web-dev to other areas of use and push its abilities to the limit.

Built With

• css
• html
• java
• javascript
• react