The history of OCR technology: from reading machines to document AI

The history of OCR (optical character recognition) starts in 1914, when Emanuel Goldberg built a machine that read printed characters and turned them into telegraph code. OCR technology then moved from 1950s reading machines to omni-font OCR in the 1970s, desktop software in the 1980s, open-source engines such as Tesseract in 2005 and deep learning in the 2010s. Today OCR is one layer of document AI, which reads, understands and checks business documents.

Below are the key milestones, with dates checked against primary or well-documented sources, then a short note on each era.

OCR history at a glance#

1. 1914: the first reading machinesEmanuel Goldberg builds a machine that reads characters and converts them to telegraph code; Edmund Fournier d'Albe builds the Optophone, which turns printed letters into tones.
2. 1927 to 1931: the Statistical MachineGoldberg applies for a patent on an electronic microfilm search machine and demonstrates it in 1931.
3. 1929 to 1935: Tauschek's reading machineGustav Tauschek files for a US patent on a "reading machine" in 1929; it's granted in 1935.
4. 1952: the first OCR companyDavid Shepard, who built the GISMO reading machine with Harvey Cook Jr., founds Intelligent Machines Research Corporation.
5. 1958: MICR for checksThe American Bankers Association adopts the magnetic-ink E-13B font for checks.
6. 1974 to 1976: omni-font OCRRay Kurzweil founds Kurzweil Computer Products; his reading machine for blind people is unveiled in 1976.
7. 1980s to 1990s: desktop OCROCR moves from dedicated hardware to software on PCs and flatbed scanners.
8. 2005: Tesseract goes open sourceHP releases Tesseract, first developed at HP between 1985 and 1994.
9. 2015: CRNNNeural networks read whole words and lines end to end.
10. 2018: Tesseract 4The open-source engine adds an LSTM neural network engine.
11. 2021 onward: transformers and document AITransformer OCR models such as TrOCR, layout-aware models and vision-language models.

Early reading machines (1914 to 1930s)#

OCR's roots are in telegraphy and in reading aids for blind people. Goldberg's later Statistical Machine, patented after a 1927 application, searched microfilmed records with a photocell: when a code on the film matched the search card, the light was blocked and the circuit signaled a match. Historian Michael Buckland calls it the first functioning document retrieval system to use electronics.

In Vienna, Gustav Tauschek's reading machine compared printed characters against templates using light and a photodetector.

The first commercial OCR (1950s)#

IBM licensed Shepard's machine but never put it into production. Reading credit cards became the first major industry use of OCR, and Shepard designed the Farrington B numeric font still used on most credit cards.

Banking and MICR (1950s to 1960s)#

Rather than teach machines to read ordinary print, banks designed characters for machines. The first MICR checks were printed by the end of 1959, the standard was almost universal in the US by 1963, and every check still carries a MICR line. See OCR vs MICR.

Omni-font OCR (1970s)#

Earlier systems read only specific fonts; Kurzweil's read text in virtually any font. His reading machine was unveiled at a news conference with the National Federation of the Blind, and a commercial version of the OCR software went on sale in 1978.

Desktop and open-source OCR (1980s to 2000s)#

With PCs and flatbed scanners, OCR became software that turned scanned pages into editable, searchable text. After HP open sourced Tesseract in 2005, Google developed it from 2006 to 2017, and OCR spread to book digitization, cloud APIs and phone cameras. For a practical guide, see Tesseract OCR.

Deep learning (2010s)#

Models stopped classifying single characters with hand-built features and learned directly from images. CRNN combined convolutional and recurrent layers to read whole lines, and Tesseract 4 followed with its LSTM engine, so accuracy on messy scans, unusual fonts and handwriting improved sharply. For the algorithms, see text recognition algorithms.

From OCR to document AI (2020s)#

Transformer models such as TrOCR (2021) pushed recognition further, layout-aware models connected text to its position on the page, and vision-language models now answer questions about a document image directly. The business question moved from "can the machine read this?" to "can it turn this document into data I can trust?"

That is intelligent document processing. Docsumo applies it to documents such as bank statements, ACORD forms and invoices, with 99% field-level accuracy on 250+ document types and fields it's unsure about sent to a person for review. See the platform or read what is OCR.

The bottom line#

In a century, OCR went from photocells and template disks to neural networks that read almost any text. The step after reading is already here: systems that understand, check and act on documents.

Book a demo with a few of your own documents, or start a free trial.

Frequently asked questions#