OmniDocWhat formats are supported for image translation and which way to scan PDF?
What formats does Image Translation support? Which route should be taken for single pictures, screenshots and multi-page scanned PDFs, how to determine whether a PDF is a scan, and what to do when editable results are needed, all explained in one go.
Key points of this article
- Single photos and screenshots will be translated into images; multi-page scanned PDFs will be processed as scanned copies. Do not mix the two.
- Judgment method: Try to select a piece of text in the PDF. If you cannot select it, it will basically be a scanned copy or picture page.
- Scanned copies can be directly uploaded for translation, and you can choose bilingual comparison or translation only. The layout and tables will be preserved as much as possible.
- When you need to change the translation, first use OCR to recognize the text, then export it to Word or modify it in a PDF editor
- When choosing a free tool, first check the file size limit, whether there is a watermark, whether the privacy terms and layout are retained
If you have a contract photo taken or a paper PDF saved with a scanner, and you want to translate the English into Chinese, the first question is often: What formats are supported for image translation? Both pictures and scans appear to be files that "cannot directly select text", but their suitable processing routes are not exactly the same. If you make the wrong choice, the layout will be messed up at best, or you will have to rework it at worst.
This article is written for readers who have pictures, screenshots or scanned PDFs and want to get the translation right the first time. After reading this, you will be able to determine whether a document should be translated as a picture or a scan, know how to quickly confirm whether a PDF is a scan, and know what to do when you want results that can be modified.
What formats are supported by image translation? One sentence conclusion
OmniDoc's document translation supports images and scanned PDFs. The judgment criterion is not just the suffix name, but also whether the file is "a picture" or "an entire multi-page document", and whether there is a text layer that can be selected.
| your situation | Recommended practices | reason |
|---|---|---|
| A taken photo or screenshot (common image formats such as JPG, PNG, etc.) | Upload directly to omnidoc.top for image translation | A single image has no page structure, the entire image can be processed |
| Scan a PDF of multiple pages, such as a scanned contract or paper | Upload the entire scanned copy for document translation | Each page is an image, and page order, layout and tables need to be preserved |
| Digital PDF with selectable text | Upload translation as normal PDF | The text layer already exists, there are fewer recognition steps, and the layout information is more complete. |
| Scanned copy, but the translation will need to be modified later | Use OCR to recognize text first, then export it to Word or modify it in a PDF editor | The scanned document itself cannot be modified directly. |
| Multiple pictures need to be translated together | First use a common method (such as system printing to PDF) to merge into a multi-page PDF, and then upload the entire copy | The page order is centralized, making it easier to compare before and after the translation. |
What is the difference between pictures and scanned copies? How to confirm?
A picture is a group of pixels with no text structure inside it. Scanned PDFs usually have one picture embedded in the PDF per page, which looks like text, but in fact, computers cannot read the characters. Before translation, the text in the picture must be read out. This step is called OCR (optical character recognition). OmniDoc's document translation supports direct uploading of scanned documents, without the need for you to manually do OCR first.
How to determine whether a PDF is a scanned copy
Open the PDF, drag and select a section of text with the mouse, and then try copying and pasting. The one that can select and copy normal text is the digital PDF. It cannot be selected, or the whole area turns gray after being selected, and the copied text is garbled or even has no content. Most of it is a scanned copy or a picture page.
Some PDFs are mixed: some pages are scanned and some pages are text. In this case, it is recommended to turn over the pages and test page by page, and then decide whether to process the entire copy as a scanned copy or separate the picture pages and process them separately.
Common misunderstandings
- Misunderstanding 1: If the suffix is .pdf, it must be treated as a normal PDF. The suffix only indicates the container format. Whether it is text or pictures depends on the selection test above.
- Misunderstanding 2: Once the scanned document is translated, it can be used as the final version. Both machine translation and OCR may make errors, especially parts that are blurry, italic, handwritten, or have small font sizes. Important content must be compared with the original text point by point.
- Misunderstanding 3: Images and scans must be converted to Word before they can be translated. Not necessary. Just upload it directly. Only when you need to continue editing, you can export to Word.
When searching for "free", look at these points when choosing a tool.
Readers who search for "pdf scan translation for free" or "picture translation for free and editable" usually want to know what tool to use. When choosing, you can first look at the following items, and then compare them with the actual instructions on the page:
- Are there any restrictions on the file size and number of pages? Can a scan of dozens of pages be uploaded in its entirety?
- Is there a watermark in the result? Will it affect submission or printing?
- How to write privacy clauses, and whether sensitive documents such as contracts and medical records are suitable for uploading
- Does the translation retain the original layout, tables and pictures, or is it just plain text?
- Whether the results can be modified further and whether the exported format meets your purpose
📊 Core comparison
| File type | Can text be selected directly? | Recommended route | Can I continue to modify the text after translation? |
|---|---|---|---|
| Single picture or screenshot | Can't. | Picture translation | The translation needs to be sorted out before it is revised. |
| Multi-page scan PDF | Usually not. | Scan the whole copy and upload it for document translation | You can first OCR, and then export Word to modify |
| Digital version of PDF | Can | Translation of ordinary PDF documents | Paragraph text can be modified in the PDF editor |
# how to use it
- First determine the type of file: a single picture is processed directly; PDF first drag and select a paragraph of text in the reader, and press the scanned piece to see if it is not selected.
two. Open omnidoc.top, upload pictures or scan PDF, and choose the translation form: bilingual control, or translation only.
- Download the results after the completion of the processing, and check the numbers, dates, names and technical terms against the original text.
- If you need to modify it, install the desktop client (download address download.omnidoc.top), open the scan with a PDF editor, do OCR, and then export to Word, or modify the paragraph text directly.
- After the modification is completed, check again whether the page number, table and picture are consistent with the original text.
# which method is suitable for you?
Just want to understand the general idea of a picture: send the picture directly and read the bilingual results. -to file or print the whole file: keep the layout, select bilingual control, and check it page by page. -continue to change words and write reports on the basis of the translation: first OCR, then export Word, and then check the layout after revision.
# frequently asked questions (FAQ)
1. What formats are supported in picture translation?
Common JPG, PNG and other pictures can be translated into pictures. The description of the upload page in the specific acceptable format shall prevail; if the format is not in the range, you can save it as a common picture format by general method, or print it to PDF before uploading.
2. Is the translation of pdf scanned parts free?
This article does not evaluate the charges of specific products, and the actual instructions on the page shall prevail. When selecting, you can focus on the file size limit, watermarking, privacy policy, and whether the translation remains typesetting.
3. Picture translation can be edited for free, how to do it?
The key to "editable" is whether the result can continue to change words. OCR the scanned piece first, then export Word, or modify the paragraph text in the PDF editor on the desktop client, which is a direct route.
4. Can you keep the form after scanning and translating?
OmniDoc's document translation retains the layout of tables, formulas, and pictures. However, complex tables are still recommended to check the cells and numbers against the original text and confirm that they are correct before using them.
5. Why is there something wrong with the recognized text?
Both OCR and machine translation can make mistakes, and blurred, slanted, handwritten or small font sizes are more likely to cause problems. Please check the amount, date and terms of the contract with the original text one by one.
# Summary
Which formats are supported in picture translation depends on whether the document is a picture or a whole multi-page document. Direct translation of a single picture, scan PDF to upload the whole copy, OCR and then export when you need to change the word. By dragging and selecting a passage in PDF, you can tell which way to go.
