ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Optical reading of universal paying form

Matej Brožič (2013) Optical reading of universal paying form. EngD thesis.

[img]
Preview
PDF
Download (7Mb)

    Abstract

    Paying bills with the universal payment order is very common in these days. It is used to pay bills in banks, post offices or at home, with the use of web banks and mobile banks on smart phones. To use the universal payment order in payment, it is necessary to read the data from it and enter the data in program for work with orders. The users of web and mobile banks and bank employees mostly do the data entering manually, by typing every character with their hands, which can be quite time-consuming task. Because the main goal of every bank is to simplify things for its costumers, we decided in cooperation with Banka Koper, to develop a system which would enable simple use of payment order. The system would be able to read the universal payment order data from its image and fill in the program for work with orders, so there would be no need of manually filling the data. This system is presented in this thesis. In the first part of the thesis we present the universal payment order and look trough some of the existing similar systems and present some of them. In the second part the development of the system and used technologies and methods are presented. In this part, the preprocessing of the image, used to find the fields we are interested in is presented. The optical character recognition of the data inside these fields is also presented in this part. The last part of the thesis looks over results that the system gives us and the future planned work and improvements.

    Item Type: Thesis (EngD thesis)
    Keywords: universal payment order, optical character recognition, image preprocessing, computer vision, edge detection
    Number of Pages: 63
    Language of Content: Slovenian
    Mentor / Comentors:
    Name and SurnameIDFunction
    viš. pred. dr. Borut Batagelj298Mentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=9995348)
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 2090
    Date Deposited: 05 Jul 2013 16:53
    Last Modified: 23 Jul 2013 12:28
    URI: http://eprints.fri.uni-lj.si/id/eprint/2090

    Actions (login required)

    View Item