Adobe Export PDF supports optical character recognition, or OCR, when you convert a PDF file to Word (.doc and .docx), Excel (.xlsx), or RTF (rich text format). OCR is the conversion of images of text (scanned text) into editable characters, so that you can search, correct, and copy the text.
When OCR is enabled, Adobe Export PDF performs OCR on PDF files that contain images, vector art, hidden text, or a combination of these elements. (For example, Adobe Export PDF performs OCR on PDF files created from scanned documents.) Adobe Export PDF also performs OCR on text that it can't interpret because the text was encoded incorrectly in the source application.
By default, OCR uses the language selected in the My Information dialog box. The OCR engine uses the selected language to interpret the scanned text. Selecting the correct language improves the accuracy of the conversion, as the OCR engine uses language-specific dictionaries for conversion. For non-Latin languages like Japanese, the OCR engine cannot interpret and convert the text unless you've selected the appropriate language.
To run OCR in Acrobat, refer the article: Edit scanned PDFs