OCR (optical character recognition) and OMR (optical mark recognition) are specialized systems that convert images on a paper to a format that is easily readable and processed by a computer. Both OCR and OMR technologies are comprised of hardware and software components. They function by reading images by scanner, which recognizes and deciphers them into an electronic form.
The benefits of OCR systems became evident in the 1950s, when a system was devised that would interpret printed messages received by the U.S. army into a readable format. The first commercial OCR system was installed in the headquarters of Reader's Digest in 1955.
The first basic OMR sensor was created by IBM in the 1930s. Everett Franklin Lindquist made and patented a successful OMR scanner in 1962. IBM developed an OMR test scoring machine and commercialized it in 1962. In 1972, Scantron Corporation made and sold OMR scanning equipment to schools to standardize the testing procedure.
OCR systems are comprised of optical character recognition software, font sets and scanners, while OMR systems include optical mark recognition readers/scanners and software.
OCR systems are widely used in the banking industry (to process checks), legal industry (to automate paper document processing and saving), health care (to process paperwork), education, government agencies and the finance sector.
OMR technology is used for a variety of applications, including the processing of educational tests, ballots, questionnaires, reports and order sheets/forms. The most commonly used application of OMR is the pencil bubble test, where students mark answers by using a pencil to darken a bubble on a preprinted sheet. The OMR software scans the document and reads the marks to automatically grade it.
OMR readers are able to input large amounts of data with minimal human intervention. They are reliable, cost-effective and fast, and they are able to handle huge volumes of data in mere minutes. An OMR scanner is able to read anywhere between 1,500 to 10,000 forms in an hour.
OCR readers can input large volumes of data into a digitalized form, which can be manipulated by a word processor. OCR systems generate fewer errors compared with manual data entry, therefore saving valuable time and costs that would otherwise be spent in purchasing error correction workstations. An OCR system is able to read 420 characters per second, with an accuracy rate of 98 percent.
OMR readers are unable to recognize machine or hand-printed characters. It is unable to electronically retrieve data since images of the forms processed are not scanned. Marks need to be within a specified area if the form is to be processed accurately.
The accuracy of an OCR system depends on the readability or the original source. OCR systems are expensive, costly to maintain and require manual intervention.
- Photo Credit barcode_02 image by Perth from Fotolia.com
What is OCR?
Optical Character Recognition (OCR) is the ability to convert a printed paper document into a document that can be edited on a...