Nobody likes receipts, what if there was a way to permanently save all your receipts and have all the text automatically parsed so you could query for keywords or other useful information?

What it does

It is a browser based application that takes a picture of your receipt and then parses the text.

How we built it

C++, HTML, CSS, Javascript, Bash, Tesseract, and willpower

Challenges we ran into

Bad receipt or image quality can drastically impact performance, so we needed to pre-process our images using tesseract and OpenCV.

Accomplishments that we're proud of

It can read your receipt for you.

What we learned

Text parsing can be very powerful.

What's next for Readceipt

Many features, uploading text(metadata) and image pairs to database, additional queries such as "how much money did I spend on gas last month?"

Share this project: