Scanned text sorted by data fields
When documents are scanned, character recognition software is used to make the text within them accessible to the computer, which enables users to work with them more effectively. Sometimes, full-text OCR is used to make the entire document searchable, while in other situations, with more complex document formats, the page can be divided into data fields so that text can immediately be extracted into the correct location based on its position on the page. Processing text in this way requires forms processing software that can recognize separate data fields and handle the text accordingly. This ultimately saves the user time that would otherwise be spent on data entry.
Recognizing handwritten text in a data field
Data fields can be designed for the entry of different types of information. Some documents, such as invoices, might consist entirely of printed text, and are fairly easy for a computer to process because the characters are fairly uniform, with only minimal variation between fonts. Other forms, on the other hand, might contain handwritten responses. Because different people have different handwriting styles, a rigid character recognition tool will not be able to identify handwritten text reliably. For the recognition of handwritten text, forms processing software must have ICR capabilities. ICR stands for intelligent character recognition, meaning that the software as the flexibility to adapt to different people's handwriting. With ICR, even handwritten responses become digitally accessible.
Reliability of handwritten text recognition
While OCR is a widely known, established tool whose accuracy is no longer questioned, some people remain uncertain about the reliability of handwritten text recognition. It is true that extremely unclear, difficult-to-read text might not be recognized correctly, just as it might not be legible to a human reader. However, in general ICR can be extremely accurate. Considering the amount of time that it can save without any loss in accuracy, it seems like an excellent addition to the document automation process. With accuracy rates that can exceed 97 percent, it is quickly becoming an indispensable tool in many offices.
[ Back ]