Payment receipts validation through duplicate elimination using optical character recognition

Abstract

Optical Character Recognition (OCR) is the method of digital image retrieval of the characters. The idea behind OCR is to obtain an image or pdf format document and extract the characters from that image and present it in an editable format to the user. This thesis focused on research related to extract the information such as vendor name, category of the receipt (food related, travel related etc.) and amount from the receipt which can be printed and hand written. Further to identifying the mentioned information, expanded the research on identifying the duplicate receipts as well. Petty Cash is an accessible store of money kept by organizations for expenditure on small items. When an employee wants to reimburse the amount that he/she spent, they need to fill a voucher with the date of the expense, amount, vendor, reason of the expense and attach the supporting documents (receipts) which will consume papers. In this digital world, easily we can automate this process using digital platforms and tools. Mobile phones are significantly playing major roles in our day-to-day life more than ever and the usage of mobile phones are increasing drastically compare to desktop computers. In order to reduce the carbon foot print, we can take necessary steps to reduce the paper usage. Building a mobile application which can automate the petty cash process which includes OCR capability on receipts would engage the users to use it in their organizations. This is the first time, OCR on receipts and duplicate identifier is researched and done. There are no researches conducted on this.