Optical Character Recognition (OCR) is the conversion of scanned images of handwritten, typewritten or printed text into searchable, editable documents. OCR software is able to recognise the difference between characters and images, and between characters themselves.
The use of paper has been displaced from some activities. For example, the vast majority of journeys on the London Underground are made using the Oyster card without a paper ticket being issued. We have witnessed talk of a paperless office for more than 40 years. However, the office environment has shown a resistance to remove the mountain of paper generated. Things have changed in the past few years, with a marked shift in the paperless office concept. Paper documents contain a wealth of important management data and information that would be better stored electronically. There is computer software that makes this conversion possible. The benefit of scanning documents is not purely for archival reasons. OCR technology is vital for gaining access to paper-based information, as well as integrating that information in digital workflows.
OCR software is not mainstream so open source alternatives to proprietary heavyweight software (such as OmniPage, ReadIRIS, CVision pdfcompressor, or the Linux supported ABBYY FineReader) are fairly thin on the ground. Matters are also complicated by the fact that OCR computer software needs very sophisticated algorithms to translate the image of text into accurate actual text. The software also has to cope with images that contain a lot more than text, such as layouts, images, graphics, tables, in single or multi pages.
Here’s our rating for each OCR system.
Now, let’s explore the 5 OCR systems at hand. For each title we have compiled its own portal page, a full description with an in-depth analysis of its features, together with links to relevant resources.
|Tesseract||High quality neural net (LSTM) based OCR engine focused on line recognition|
|ocropy||Open source document analysis and OCR system|
|Cuneiform||OCR Engine to convert OCR documents into editable form|
|Ocrad||OCR engine based on a feature extraction method|
|GOCR||Reads images in many formats|
|Return to our complete collection of recommended free and open source software including our latest additions.|