Document separation is the means by which document capture software splits images as they are scanned. This is one of the key automation features in any capture application, and allows you to take advantage of the large document feeders on scanners and “scan the stack”. For example, with an AP Invoice Processing solution, you can take a group of 30 invoices, and the scanning software will find when one stops, and another begins. Methods of document separation are as follows:
- OCR Separation – this method uses Optical Character Recognition (OCR) to find key terms on the first page of a document, and then split when a match occurs. This typically requires a pattern matching, or advanced extraction engine to accomplish this task. OCR can be refined further through the use of zones to increase the speed and improve accuracy.
- Page Separation – the simplest of all methods, this simply counts pages before a new document is created.
- Folder Separation – this method combines all the pages within a network or PC folder into a single document.
- Pattern Separation – this is an advanced method of separation that recognizes patterns and/or images on the first page to identify when to split documents while capturing.
- Hybrid Separation – this method combines several of the above methods to minimize false separation and maximize accuracy when the document capture and scanning process is running.