How to convert scanned PDF forms to Excel: OCR first and then export

Key points of this article

First try to select the text in the table. If you cannot select it, it is usually a scanned copy and you need to do OCR first.
Scanning quality directly affects the recognition results. If the page is skewed or blurred, you should rescan it first.
After exporting, check the numbers, decimal points, negative signs and dates against the original table. This type of content is easy to misread.
Merged cells and complex headers may require manual adjustment in Excel

If the PDF you have is a scanned copy, such as a financial statement generated by taking a photo, a paper questionnaire, or a contract attachment, and you want to put the tables into Excel for sorting and calculation, directly copying and pasting will often only result in a bunch of misplaced text, and sometimes nothing can be copied. The reason is that the tables in this type of PDF are essentially pictures, and there is no text layer that can be directly extracted. The conventional "extract table" idea does not work.

This article is written for people who need to convert scanned PDF tables into Excel. The core method is divided into two steps: first use OCR to recognize the text in the picture, and then export the recognition results to an Excel table. After reading, you will know how to determine whether a PDF is a scan, how to complete the identification and export, and how to check whether the data has been read correctly after exporting.

How to convert scanned PDF tables to Excel? One sentence conclusion

Let me start with the conclusion: Scanned PDFs cannot directly export tables. You must first use OCR to turn the images into recognizable text, and then export them to Excel. If the PDF itself has a text layer, you can skip OCR and export it directly.

your situation	Recommended practices	reason
Table text can be selected and copied in PDF	Export directly to Excel	The file already has a text layer and does not need to be recognized
Scan or take photo with mobile phone to generate PDF	First OCR recognition, then export to Excel	The table is a picture and must be turned into text first.
Table spans multiple pages	Arrange the order of pages first, and then identify and export them uniformly	Paging processing can easily cause table header misalignment
There are a large number of amounts or numbers in the table	After identification, check item by item, focusing on numbers.	OCR is more likely to misread similar glyphs and decimal points
Just need a few pages	Split the page containing the table first and then process it	Narrowing the scope reduces the workload of reconciliation

Before you start: Confirm these points first

Confirm whether the PDF is a scanned copy. Use the mouse to drag a number in the table. If it can be selected and highlighted, it means there is a text layer; if it cannot be selected, it can basically be judged that it is a picture. This step determines whether you want to do OCR, which can save unnecessary operations.
Check the picture clarity. When the scanning resolution is too low, the page is skewed or has shadows, the recognition results will be significantly worse. Try to place the photo as straight as possible and with even lighting; for pages with blurred text edges, try to obtain a clearer version before identifying them.
Determine which ranks you want to transfer to. Table headers, merged cells and slash headers will affect the export effect. Read through it first and decide whether to transfer only the data area or the table header together.
Prepare the environment to open Excel. The exported file is in the common Excel table format and can be opened using desktop office software or online forms. If you need to do formula calculations later, you can export them and then organize them in a unified manner.
Allow time for verification. OCR is machine recognition and cannot ensure that every character is correct. Important data such as amount, number, date, etc. need to be compared item by item with the original form.

Steps to convert scanned tables to Excel using OmniDoc desktop client

Download and install the desktop client. Go to download.omnidoc.top to download the version corresponding to your system (Windows, macOS or Linux), install it and open it.
Open the scanned PDF. Open this document in the client, turn to the page containing the table, and confirm that the table is complete and the page is not obviously skewed.
Perform OCR on the scanned page. Use the OCR feature in the PDF editor to recognize text. If you only need some pages, you can split these pages first and then identify them to reduce the waiting and verification scope.
Check the recognition effect. After the recognition is completed, try to select the text in several cells and compare it with the original image. If an entire line of bits or text is missing, the scan quality may be insufficient. It is recommended to go back to the previous steps to obtain a clear page again.
Export to Excel. Select Excel format in the export option, export the recognized file, and save it as an .xlsx file.
Open the exported table and check the structure. Check whether the number of rows, columns and headers are consistent with the original table. Complex structures such as merged cells and slash headers may require manual adjustment in Excel.
Check the data item by item. Focus on comparing numbers, dates and amounts. Punctuation, negative signs and decimal points are common misunderstandings.
Save after correction. Modify the wrong cells directly in Excel and save it as the final file after confirming it is correct.

Check each item after completion

Check items	How to see	What to do if there is a problem
Is the text complete?	Compare with the original page to check for missing characters or missing lines	Returning to the OCR step, switch to a clearer scan and re-identify
Whether the rows and columns are aligned	Check if each row of data falls into the correct column	Manually split or merge in Excel, and export in segments if necessary
Numbers and symbols	Check 0 and O, 1 and l, 6 and 8, decimal point, minus, and thousandth comma	Compare each table with the original table one by one, correct it, and then perform summing and other calculations
The header and the total plan proceed	Check whether the meter header is complete and whether the total matches the original table	Recalculate the total using Excel formulas and compare it with the original table
Long number string	Check whether your ID number, order number, bank card number, etc. have been altered	Manually correct based on the original table, and set these columns to text format

❓ Frequently Asked Questions (FAQ)

1. Can table photos taken on a phone be directly converted to Excel?

You can first organize the photos into PDF or image files, then follow the steps above for OCR recognition and export. When taking photos, try to position the photo correctly and ensure even lighting; the recognition results will be more reliable. Photos taken with skewness or blurry will have a noticeably higher rate of recognition errors.

2. Why do the recognized tables become misaligned?

Common causes include merged cells in the original table, unclear table lines, or skewed pages. First, check the scan quality, and reacquire the page if necessary; After exporting, adjust rows and columns in Excel.

3. What can be done when there is no OCR function?

Key data can be entered manually or using online conversion websites that recognize forms. Note: Do not upload contracts, financial documents, or other sensitive documents to websites of unknown sources. Check the privacy policy before uploading.

4. After conversion, can the original table font and color be preserved?

Fonts, colors, and detail styles are usually impossible to fully retain, so the focus should be on whether the data is correct and the structure is clear. For neat layout, export and organize in Excel.

5. What should you look for when choosing free tools?

Pay attention to file size limits, whether exported results have watermarks, whether the table structure can be preserved, and whether the privacy policy explains how files are stored and deleted. These few items determine whether the result can be used even more than the word "free."

📝 Summary

The key to converting scanned PDF tables to Excel is to first convert images into text using OCR, then export and verify each item. Accurately identifying the PDF type and ensuring clear scanning can save most of the rework time. For important data such as amounts and numbers, be sure to compare them with the original table before use.

Open OmniDoc and process your table file