Building a Receipt Parsing Web Application

Dátum
Folyóirat címe
Folyóirat ISSN
Kötet címe (évfolyam száma)
Kiadó
Absztrakt

Modern consumer expenses-tracking software is generally not the most useful or developed. Paper receipts are cumbersome and it is difficult to track itemized costs from them manually. If the content of a receipt could be read by software, and individual purchases parsed and recorded, expenses could be tracked and calculated much more easily, and paper receipts and budgeting software could be disposed of. To achieve this, a web application capable of parsing and structuring specific data from a receipt was designed and developed for this thesis. A simple web API was built, with optical character recognition and image processing technologies employed to prepare and read data from images of collected receipts. Data collected is fed through a parsing algorithm written to understand the information layout of each receipt, and collect the relevant values. The application is largely a success, producing readable and reliable data, but the web application needs further development for full usability. Possible further improvements for the parsing algorithm are discussed.

Leírás
Kulcsszavak
OCR, Optical Character Recognition, Web Application, Web Development, Image Processing, API
Forrás
Gyűjtemények