Information retrieval and processing from printed invoices with Java technology

OData support
Dr. Lengyel László
Department of Automation and Applied Informatics

The printed receipts take a decisive part in our daily lives. Most of the time after purchasing an item we still receive the document printed, or rarely manually written, on paper. However, digitization provides a much clearer and more secure way to store the documents. This is another challenge for us to handle the digitally generated receipts together with the printed ones that we have already issued.

The main idea of the dissertation is an application which is capable of digitizing printed receipts, gathering important information and displaying them through a user interface. This provides a service for the ordinary person with which he can have his bills and thus his finances, maintained, digitized and managed.

In the dissertation, I implemented a software that automatically generates a digital receipt, including the important information, with optical character recognition, after uploading the scanned version of it. The user can view, manage, and modify the digital receipt through an online interface.

From technological side I developed the system in Java, using the Spring Framework and storing the data in a HSQL database managed through the Hibernate library.


Please sign in to download the files of this thesis.