What is OCR?
OCR = Optical Character Recognition. Optical Character Recognition involves reading text from paper and translating the information into a computer friendly format (for example, into ASCII codes). An OCR system enables you to scan text data from documents such as Passports, Visas and Travel Documents; Invoices; Forms, etc. directly into a computer file for digital editing, manipulating, and processing. All OCR systems utilize an optical scanner for reading text, plus sophisticated software for analyzing images. Most systems use a combination of hardware (specialized circuit boards) and software for character recognition. Advanced OCR systems are able to read text in a wide variety of styles and fonts.
“OCR” stands for optical character recognition. The scanned image pages of the documents on Profiles in Science are subjected to OCR to obtain the text within them. An example of a scanned document and its corresponding OCR text can be found atFSBBDL. OCR is often imperfect, particularly for historical documents such as those found on Profiles in Science. OCR is ineffective for handwritten documents. Profiles in Science is currently making the OCR text available as an experiment. In cases where the OCR text is accurate, it may be helpful to someone with low vision who is using a screen reader. No OCR text is listed for scanned documents that have already been transcribed since the transcript is more accurate than OCR. If you have suggestions for making the digitized materials on Profiles in Science more accessible for users with vision problems, please email them toprofiles@nlm.nih.gov.
A. OCR stands for Optical Character Recognition. It refers to the process by which a scanned document is electronically analysed, so that the text information can be extracted from the image and reproduced as a text file. Text generated by OCR is often input into text search databases, allowing retrieval of the original scanned image based on its content. This text information can also be embedded in your PDF or Word file to make these documents fully searchable as if they were pure text files.
