For quick document indexing or simple forms processing, you may be willing to key information manually from the image, but in most capture applications, the goal is to automate as much as possible using technologies that automatically read the information on the page. These are called recognition technologies.
Different types of information require different data recognition technologies. A specific recognition technique from a vendor is often called a "recognition engine". Some engines, known as ICR, recognize hand printed characters, while others, called OCR, specialize in reading typed characters; still others read barcodes, and OMR engines read check boxes. Choose the right engine for the job and you'll get optimal productivity.
There are vendors who specialize in recognition engine technologies. Most vendors provide a sub-set of the full range of engines available. Learn more about recognition engine vendors.
Recognition engines alone are raw tools. They convert images into characters, provide useful information about the success of the recognition, the recognition "confidence" on a character-by-character basis, but they generally don't understand much about the meaning of the information. For example, the engine may be told to only recognize numbers, but to know the validity of the value a calculation or lookup may be required. That's usually handled by the capture software that controls the recognition engine. Learn more about Datacap's Recognition.
The most accurate manual data entry requires having two people type the data and then to have the results compared. The same principle works with recognition: use two different recognition engines and combine their results in a process called "voting" to select the best candidate value for a field. Voting can dramatically reduce the amount of manual review required while greatly increasing accuracy.
back to top | request more information | contact Datacap | site map