Automatic invoice interpretation: invoice structure analysis
D.A. Kosiba, Rangachar Kasturi · 1996
We propose a method of invoice document structure analysis that provides a means to extract the relevant information from an unknown invoice. Our method uses a combination of textual and graphical processing by analyzing the line and line intersection features in the document as well as searching for possible keywords such as item number, quantity, total, etc. Valid keyword search regions are determined by a specialized connected-component analysis before any OCR is performed. The results of the the keyword search and the line analysis are combined to give the search regions for extracting the relevant data contained in the invoice. This analysis will become part of a larger invoice interpretation system which is currently under development.