Recognise and Classify

  • ELO DocXtractor


[Return to Top] 


ELO OCR’s task is to analyse and convert graphical information into text information.

ELO OCR enables the user to extract any text information from image files (tiff, pdf, jpg, etc.). Once a document has been scanned in (e.g. an invoice, order or delivery note), OCR reads the necessary information (customer number, address, amounts, etc.) and automatically imports this into the keywording.

If a company uses standardised forms (orders, contracts, invoices, delivery notes), ELO OCR can implement automated document archiving with keywording. Hence, there is no need for manual processing. Another big advantage of ELO OCR is the full-text search of TIF documents. This makes it possible to search all of the content of scanned documents and not just the keywording index. This ensures directed and quick searching.

[Return to Top]

ELO DocXtractor

ELO DocXtractor is an application for the intelligent and automated classification and processing of incoming documents. The system trains and learns both the structure and identifying features on the basis of a learning set with a small selection of sample documents. Subsequent to this, it is then in a position to analyse the content of new documents also, to evaluate this and allocate it with a very high degree of accuracy on the basis of classes.

Thanks to the pre-defined configuration models, ELO DocXtractor can be adapted to meet the specific requirements of your company. The smart blocks encapsulate self-contained functional blocks such as table capture, order allocation and payment practice. These can then be individually activated for each customer. The benefits: simple to use, transparent documentation and lean client configuration.

For example, ELO DocXtractor recognises whether an incoming fax concerns an order or an invoice and automatically extracts the data needed in each case, such as the invoice number, amount, date or individual invoice positions. All of the data is then classified. If the fax is an invoice, the module checks whether the right orders or delivery note information are also available in the ERP system. Furthermore, the system is able to automatically supplement incomplete data and, for example, update a missing customer number from the existing customer database. ELO DocXtractor also recognises and captures hand-written additions to machine-printed documents.

In this way, business documents and their contents can be transferred almost completely automatically to the relevant processes and the practice of manually entering invoice data into the accounting system becomes a thing of the past. One area in which ELO DocXtractor can be used, for example, is the intelligent processing of incoming invoices in ERP systems such as SAP and Microsoft. When used for this, the module automatically classifies all incoming invoices and allocates them to the right suppliers. This leads to high recognition rates which ensure the quick development of considerable potential for rationalisation. This results in an exceptional return on investment (ROI) and often a payback period of less than one year.