|
Document Splitting and Matching
When scanning a batch of documents in one scan run, two requirements need to be handled - splitting of the batch into individual documents and matching of those individual documents against a definition.
For example, scanning a batch of invoices involves splitting the batch into individual invoices and then matching each invoice to a supplier so that the document can be analysed further.
SPLITTING
Splitting of document batches involves teaching Scan2x what to look for in order to determine how to split. Examples of splitting conditions are:
•A splitter page - this is a dedicated, reusable document that is uniquely identifiable to Scan2x as the differentiator between one document and another in the batch.
•A Barcode or QRCode on the document itself. If each document in the batch has a barcode or QR on its first page, Scan2x can detect and use this.
•A number of pages (e.g., split every three pages)
•The presence of a value in a specified place on the page
MATCHING
The recognition of a document as being linked to one entity, or another can be done in two ways.
1.Fingerprint recognition - the ability to recognise the structure and layout of one document versus others in order to make a best-guess decision as to which document type it belongs to. Fingerprint recognition (or ADR - short for Automatic Document Recognition) is described in more detail under the Document Matching (ADR) heading.
2.The presence or otherwise of a particular value in an indicated zone on the page. By creating rules to positively identify a document, Scan2x will be able to differentiate between different document types. See the Document Splitting Rules section for further detail.
|
