OCR Software and Application Features
We typically use the term OCR (Optical Character Recognition) to mean the conversion of an image to text. There are several OCR functions that are utilized in advanced document capture software, such as PSIGEN PSI:Capture, and your every day scanning software, such as VizitSP.
For an in depth definition, go here -> Wikipedia OCR Software
So let’s talk a bit about the different OCR functionalities:
Full Text OCR
Full Text OCR takes the entire image and converts it to a text output. The OCR output can be in several formats, including: Searchable PDF, Microsoft Word, Plain Text, HTML, etc. The main goal of full text OCR is typically “searchability”, and the results are usually placed into a backend repository, such as Microsoft SharePoint (For more about SharePoint Scanning and Capture – SharePoint Scanning).
Zone OCR only looks at a particular region, or zone, of the scanned page and converts just that portion to text. There are several reasons to use zone OCR rather than full text:
• You only want to search on the information in that zone.
• Full Text OCR takes much longer, so Zone OCR can speed up processing time.
• You want to extract the contents of the zone, and place it into an index field.
Most advanced document capture applications provide the ability to map the contents of a zone to an index field, that can then provide granular search capabilities based only on that field.
OCR-Assisted Indexing, or point-and-click indexing, provide the user the capability to just click on words or segments of a document, and convert that image portion to text. This capability exists in many different capture applications, and provides a simple, easy indexing function on documents.
Rubberband OCR provides the ability to drag a box with the mouse over a portion of text, and automatically convert that segment into text, and even place it into an index field. It is similar to OCR-Assisted Indexing, but allows the capture of large portions of text on a scanned image.
One of the key challenges in document scanning and capture is the ability to easily split a stack of paper into individual documents. Advanced Document Capture Software can provide the ability split whenever a key term or word is found on a page through the OCR process. Utilizing the Optical Character Recognition engine in this manner can save on document preparation time before scanning and capture.
Advanced Data Extraction
Many of the Document Capture applications on the market today provide a means of extracting data through some type expression matching, or extraction engine. OCR Software is utilized to do the text conversion prior to the extraction. For an example of data extraction, see this YouTube Video - EOB Processing and Data Extraction.
Another key use of the OCR results can be the identification of documents. Optical Character Recognition can identify key elements on a document, and then determine how to process it based on those elements.