OmniDocSteps on how to convert scanned documents into Word documents
Scans are images and cannot be edited directly. This article explains the specific operation of converting scanned documents to Word according to the four steps of "confirming the file type, OCR recognizing text, exporting to Word, and checking item by item", and explains how to choose for different purposes.
Key points of this article
- First try to select a piece of text in the PDF. If you cannot select it, it means it is a scan or image and needs to be OCR recognized before exporting.
- After OCR recognition, you must first check whether the text can be selected and copied before exporting, otherwise the exported Word may be just a picture.
- The exported Word must be checked against the original page by page, with numbers, amounts, names of people and professional terms being the key checking objects.
- If you just want to copy text and if you need to keep the layout and continue editing, the methods are different. Choosing the right path can save you rework.
Many contracts, certificates, old papers and paper documents only have scanned copies. Scanned documents are essentially pictures, and Word cannot directly modify the text in them. What is copied and pasted is often a bunch of garbled characters or blank spaces. If you want to turn it into an editable Word document, the core is to first recognize the text in the picture and then export it to Word.
This article is written for office workers, students, and small office users who need to edit the content of scanned documents. After reading, you will know how to determine the file type, how to use OCR to turn scanned documents into editable text, what areas to check after exporting, and which method to choose for different purposes. The following steps take OmniDoc's PDF OCR and export functions as an example. The general idea applies to other tools as well.
How to convert scanned documents into Word? One sentence conclusion
Directly insert the scanned document into Word, and all you will get is a picture, and the text still cannot be changed. The correct sequence is to perform OCR recognition first, then export to Word, and finally verify. The following table provides recommended practices based on common situations.
| your situation | Recommended practices | reason |
|---|---|---|
| The scan is a PDF, just copy a few paragraphs of text | First OCR recognition, then select copy | No need to export the entire file, fast, suitable for partial reference |
| The entire document is required to continue editing. | Export to Word after OCR recognition and modify in Word | Word paragraphs, titles and tables facilitate subsequent typesetting |
| The original is a paper contract or certificate, and the electronic version must be kept | Keep the original scanned PDF as a voucher and save a copy in Word for editing | The original is used for documentation and the edited version is used for revision. It is safer to keep the two separate. |
| The file contains a large number of tables or formulas | After exporting, focus on checking the tables and formulas and manually correct them if necessary. | Recognition tools may be biased in restoring complex layouts |
| Only Chinese translations of foreign language materials are needed | Use the document translation function to process it, and then export it as needed. | Translation and format conversion can be done together, no need to separate them in two steps |
Before you start: Confirm these points first
Make a few judgments before taking action to avoid detours.
- Confirm that the document is really a scan. Open the PDF and try to select a piece of text with the mouse. If it cannot be selected, or if it is a whole picture after being selected, it means it is a scanned copy. If you can select text, you can export it directly to Word without OCR.
- Check scan clarity. If the text is blurred, skewed, shadowed or stamped on the text, recognition errors will increase significantly. Lay the paper flat and adjust the brightness before scanning, and the effect will be much better.
- Clear purpose. If you are just excerpting a few paragraphs, you do not need to export the entire copy; if you need to keep the format and continue to modify it, you must do a complete export and typesetting check.
- Pay attention to privacy. For sensitive documents such as contracts, medical records, and identity certificates, you should give priority to using a desktop client that can be processed locally, and confirm that you have the authority to process the document.
- Have original documents ready. When checking, you need to compare it page by page with the original. Only by retaining the original scan can you determine where the identification errors are.
Steps to convert scanned documents to Word in OmniDoc desktop client
If you want to complete OCR and export locally, you can follow the steps below. OmniDoc's desktop client supports Windows, macOS, and Linux, and the download address is download.omnidoc.top.
- Install and open the desktop client. Download the version corresponding to the system from the official website, install it and start it.
- Open the scanned PDF. Drag the files that need to be converted into the client or import them through the menu, and confirm that the page can be displayed normally.
- Perform OCR recognition on the file. Use OCR in the PDF editing function to let the software recognize the text in the image. When there are many pages, wait patiently for the recognition to be completed before proceeding to the next step.
- Check whether the text can be selected. After the recognition is completed, try to select a piece of text and copy it. If it can be selected, the description text layer has been generated; if it cannot be selected, you need to recheck whether the OCR execution is completed.
- Fix obvious recognition errors. If there are typos or sentence breaks in individual paragraphs, you can directly change the paragraph text in the PDF editor before proceeding to the next step.
- Export to Word format. Select Export to Word and save to a local folder. It is recommended that the file name be marked with "edited version" to distinguish it from the original scan.
- Open in Word and read through. Confirm that the file can be opened normally, and then browse it from the beginning to see if the structure of paragraphs, titles, and tables is basically reasonable.
If the exported document still needs to be collaborated online or typed in Word style, it can also be opened in app.unidoc.top’s UniDoc web editor, which supports Word-compatible editing.
Check each item after completion
After you get the Word file, do not send it out or sign it directly. Check it according to the table first.
| Check item | What do you think? | What if there's a problem? |
|---|---|---|
| Whether the text is complete or not | Check the original page by page to see if there are any missing paragraphs or lines. | Go back to PDF to check whether the page OCR is complete, and re-identify the page if necessary |
| Number, amount, date | Focus on checking the contract amount, ID number, date and number | Manual correction shall prevail with the original, and this kind of content should not be based solely on the identification results. |
| Names of persons, place names and technical terms | Check to see if there are near word errors | Replace it uniformly and use Word's lookup function to make sure there are no omissions |
| Paragraphs and headings | Check whether the title levels, line breaks and paragraphs are in line with the original text | To reformat paragraphs in Word, you don't need to go back to the scanner and redo it. |
| Form | Check whether the rows and columns are aligned and whether the contents of the cell are serial. | Complex forms can be rebuilt manually, and simple forms can be modified directly. |
| Seals, signatures, handwritten notes | To see if these contents are identified as random words or missing. | The recognition effect of handwritten content is usually limited, so it is recommended to keep the original picture and insert it into the corresponding position. |
Machine identification and export can make mistakes, and all contents involving amount, time limit, responsibility and identity information should be manually checked with the original before use.
# which method is suitable for you?
- only want to copy a few paragraphs of text: do OCR identification for the page you need, and select copy instead of exporting the whole file. - the whole document needs to be revised: complete OCR and export, page-by-page revision in Word, and complete check. - to keep the electronic version of the paper contract: the original scanned copy is saved as a voucher and the edited manuscript is saved, and the two documents are distinguished by different file names. - Foreign language scanning materials need to be read in Chinese: use the document translation function first, then export as needed, and check the technical terms.
# frequently asked questions (FAQ)
1. Can the photos taken by the mobile phone be directly converted to Word?
You can first organize the photos into a clear PDF or picture file, and then do OCR recognition. When taking pictures, try to make sure that the light is uniform, the paper is flat, and the text is not obscured, so that the recognition effect will be more stable.
2. How to convert scanned pieces into Word documents for free?
When choosing a free tool, first look at the file size limit, whether the exported file is watermarked, the privacy instructions after the file is uploaded, and whether typesetting can be retained. Try to choose the way that sensitive files can be processed locally, and do not upload them to unfamiliar websites at will.
3. The exported Word typesetting is very messy, what to do?
Typesetting disorder is mostly caused by complex layout or skewed scanning. First unify the paragraph and font size in Word, and then check the table and column; if there is a serious mess, you can only retain the text content and rearrange it according to the original.
4. Can you recognize the handwriting in the scanned piece?
The recognition effect of handwriting is usually unstable, and the results need to be checked manually. If the handwritten content is important, it is recommended to keep the original picture, insert the appropriate location in the Word, and mark the corresponding content next to it.
# Summary
How to convert scanned parts into Word documents, the key is to first confirm the file type, then use OCR to turn the text in the picture into selectable text, and finally export and check item by item. By following this set of steps, you can get an editable Word version even for old paper materials.
