At it's conception these tools will parse an image of a receipt into text. Later iterations of these tools will train machine models to parse, organize, label, and tag receipts for tracking grocery prices over time and by location. I set out to do this with R but there may be occassions where Python is necessary.
- Scan receipt by taking an image with a smartphone.
- Upload image to server that runs analysis script
- Machine model dumps results of analysis into a temporary form
- User checks form and submits data to database
- Simple Process Server (run locally?)
- Analysis script
- Build a form
- Build DB, data tables