Iterative Processing (OCR): Difference between revisions
Configadmin (talk | contribs) Created page with "Iterative processing improves OCR accuracy by performing two OCR passes, which will increase the likelihood of capturing image elements not recognized by the initial [..." |
Dgreenwood (talk | contribs) No edit summary |
||
| Line 1: | Line 1: | ||
Iterative processing improves [[OCR]] accuracy by performing two [[OCR]] passes, which will increase the likelihood of capturing image elements not recognized by the initial [[OCR]] pass. On the second pass, characters recognized in the first pass are digitally dropped out of the image, after which [[OCR]] is performed again on the remaining characters. This improves the efficiency and accuracy of recognition of elements not captured on initial processing. | Iterative processing improves [[OCR]] accuracy by performing two [[OCR]] passes, which will increase the likelihood of capturing image elements not recognized by the initial [[OCR]] pass. On the second pass, characters recognized in the first pass are digitally dropped out of the image, after which [[OCR]] is performed again on the remaining characters. This improves the efficiency and accuracy of recognition of elements not captured on initial processing. | ||
== See Also == | |||
* [[Synthesis]] | |||
Revision as of 11:00, 18 September 2020
Iterative processing improves OCR accuracy by performing two OCR passes, which will increase the likelihood of capturing image elements not recognized by the initial OCR pass. On the second pass, characters recognized in the first pass are digitally dropped out of the image, after which OCR is performed again on the remaining characters. This improves the efficiency and accuracy of recognition of elements not captured on initial processing.