SAP OCR Software

The OCR software application starts with an OCR-project (which includes the customer-specific recognition themes settings). After that, various data (e.g. invoice date, IBAN, order number, delivery note number etc.) should be extracted from a document using the freeform-text recognition-module inside the application. In case the process fails entirely or partially, the document can, however, be forwarded automatically or manually to the OCR training-workspace.

In general, The OCR-training module is an optional component of the inbox processing server biz²ScanServer, which enables the position training of different values based on different index fields. The index field positions determined for the respective document type during the OCR training are thereafter saved in a database. As a result, this information remains available when you work with a next document of the same type. Besides, now the OCR application knows where to look for the respective information inside the document.

Short overview of the documents „OCR training“ process (invoices, delivery notes, etc.):

In case a document “training” needed, it takes place in a „training workspace“.

• The training of a document does not necessarily include all index fields, as certain documents types could possibly lack them.
• During the training you indicate with the help of your cursor, where each field that needs to be trained is located in the document.
• Since all the recognition results for every document are saved in the OCR training, you can simply indicate by a mouse-click, where the data relevant for a certain field are stated.

There are two ways of submitting documents to OCR training: The OCR software includes a module which determines whether a document needs to be trained. Additionally, the validation-user is given the possibility to explicitly forward documents to the training, if he is unsatisfied with the recognition results.

Different optimization strategies: the OCR training can be completed either in advance, or during the production process.

Furthermode, to receive the best results from both modules (freeform text recognition and OCR training module), the OCR software additionally has a “voter”.  As a matter of fact, it compares and evaluates the results from both modules. Thus, the final result is a combination of both modules, which ensures a higher and better text recognition ratio than the freeform text recognition module alone.