An old scanned procedure should be possible to revise
Recover the text structure of a printed document from its scanned image. Markdown makes the result easy to edit, but page furniture, multi-column layouts and footnotes need attention before the document reads in its intended order.
Maintaining image-only procedure documents
Recover headings and numbered steps from an old scan so you can revise a checked text copy and preserve the original for reference.
Preparing source excerpts for research notes
Bring visible passages into Markdown, checking quotation accuracy and adding bibliographic information yourself.
Recover the page's reading order before editing its structure
- Use an upright scan image with the entire relevant text block visible.
- Review heading hierarchy and reading order, removing repeated page furniture when appropriate.
- Edit footnotes and page joins in your Markdown editor while comparing with the source document.
Multi-column scans, footnotes and PDF inputs
- Does document image to Markdown accept a complete PDF?
- This page accepts images of document pages. Use the appropriate PDF text workflow for an existing PDF file rather than renaming it as an image.
- Will two-column pages be read in the correct sequence?
- Columns can be interleaved by recognition. Check the reading order, or process clearly separated text regions when the layout is complex.
- Are footnotes automatically connected to their references?
- Do not assume the relationship is rebuilt. Check footnote markers and add the syntax your Markdown editor expects.
- Does OCR modernize or summarize the document wording?
- Use this workflow to recover visible content, then review it against the source. Rewriting or summarizing is a separate editorial task.
- How do I keep the page order when converting a long document?
- Name the page images with numbers, such as page-01 and page-02, convert them in that order and paste each result under the previous one. Keep a note of the first and last line of each page so you can check the joins, where a sentence continues across a page break.
- Why do running headers and page numbers appear in the text?
- They are printed on the page, so they are read like any other line. Delete them as you join the pages, or they interrupt sentences. Their position at the top or bottom of each result makes them easy to find by scanning the first and last lines of each page.
- Can I convert a document with tables and text on one page?
- Yes. Words that line up in columns are rebuilt as pipe tables with a header row, and the remaining text arrives line by line. Check the table cells and the position of each table relative to the text, since the reader works from where words sit on the page.
- Will the Markdown show bold headings and italics from the document?
- No. Emphasis and font sizes are styles in the picture, so headings arrive as ordinary lines. Mark them with # signs and add emphasis marks where the printed document used bold or italics, and keep the scan to see which lines were headings.
- Can I process an old, slightly faded scan?
- Faint print is read less reliably, and missing strokes cannot be reconstructed with certainty. Rescan at a higher quality if the original is available, keep the page flat and evenly lit, and check every amount, date and name against the scan.
- Is Markdown the right format for a document I will keep editing in Word?
- Markdown suits wikis, notes and repositories. If the destination is a Word document, Image to Word can place recognized paragraphs in a .docx, and you can start from there. Choose Markdown when you want plain text that opens in any editor.
