Automatic Document Classification
Minimize or Eliminate Pre-Sorting with CCADC
CAPSYS CAPTURE ADC™ (CCADC) is technology designed for automatic document classification, recognition and categorization tasks in a data and document capture system. CCADC allows your CAPSYS CAPTURE applications to quickly identify invoices, checks, forms, orders, delivery notes, page separators or any kind of structured document.
CCADC technology delivers a wide range of document recognition functions that can be integrated into a variety of applications including: scanning, archiving, indexing, sorting, classification, search, ECM, line of business systems or proprietary information management systems.
With CCADC you can automatically or manually assign an electronic document to one or more categories, based on its contents. The result is less document preparation and faster processes.
Text and Image Based Document Classification
CCADC utilizes two primary forms of document classification technology: Text based and Image Based Classification. Each method has its own distinct advantages depending on the application requirements that are desired to be achieved providing our customers with a high degree of configuration flexibility.
Text Based Classification allows for keyword(s) to be detected in a form in order to determine its appropriate document type. The keyword(s) can be defined as required to match expected text precisely or can be configured to be required to meet the sampling with a "degree of confidence."
Image Based Classification allows for a batch sampling of images that may slightly vary yet be grouped together and are subsequently assigned to a document type or document class. What is the benefit of having multiple samplings added to a single document type or class? Adding multiple sample images to the same class can increase the overall confidence of a recognition process. For example, such an approach allows you to "merge" 2 very similar templates in one such as an invoice that includes "VAT" and another without "VAT."
These aggregated samplings will be used to compare any new image introduced to the system against the library of known samples - allowing the CCADC module to automatically assign the proper document type to the incoming image(s). Samples may be automatically added to the Classification engine so the CCADC engine continues to learn and refine its confidence levels as new form variations are introduced into the CCADC library.