Optical Character Recognition

If the character recognition module is installed, documents can be scanned in lineart mode, saved in the proposed format, then processed for text recognition from the Image menu. Process the entire document or only a previously selected area. A configuration dialog tells the module whether the original text is in printed type, handwriting, or standardized type. Also set the language so the module can process the document correctly. See Figure 22.5: “OCR with Kooka (Optical Character Recognition)”.

Figure 22.5. OCR with Kooka (Optical Character Recognition)

OCR with Kooka (Optical Character Recognition)

Switch to the OCR Result Text window and check the text, which may need to be proofread. To do this, save the text with File -> Save OCR Result Text. The text can then be processed with OpenOffice or KWrite.