Snap a photo or take a screenshot of a printed or on-screen table and turn it into a real spreadsheet. The text is read with OCR, split into rows and columns automatically, and shown in an editable grid so you can fix any cell before downloading an Excel or CSV file.
Last updated:
Sample table
Best results: a straight, well-lit photo or a screenshot, with the table filling most of the image. Crop away anything that is not part of the table.
Spreadsheet
Press Extract table to read the image. You can also type straight into the grid or paste cells from another spreadsheet.
Advertisement
Advertisement
How to use the Image to Excel
Add your image: drop it, choose a file or paste a screenshot with Ctrl+V. A sample table is loaded so you can try it right away.
Pick the language of the text and leave Columns on Detect automatically.
Press Extract table. The first run downloads the OCR engine and language data, then the grid fills in.
Check the grid: click any cell to fix it, use Add or Delete row/column, or set an exact column count and extract again.
Download Excel (.xlsx) or CSV, or press Copy for Excel / Sheets and paste.
How rows and columns are detected
OCR returns each word with a bounding box. The converter then rebuilds the table in three steps:
Grid lines. If the table has printed borders, long dark horizontal and vertical lines are found in the image and used as row and column boundaries. This also keeps multi-line cells together.
Rows. Without horizontal lines, words whose vertical centers are within about half a line height of each other form one row.
Columns. Words that sit close together become one cell. Then the tool looks for vertical strips that almost no row touches: those empty strips are the gaps between columns. When you set a column count, the widest gaps are used.
Getting a clean result
Problem
Likely cause
Fix
Two columns merged into one
Narrow gap or a wide header spanning both
Set Columns to the real count and extract again
One column split into two
Long cell text with big spaces
Set the column count, or delete the extra column in the grid
Rows mixed together
Tilted photo
Retake the photo straight on, or rotate and crop it first
0 read as O, 1 read as l
Font or low resolution
Keep Store numbers as numbers on; it fixes these in number cells
Missing or garbled words
Blur, glare or tiny text
Use a higher-resolution image with even lighting
Tips
Screenshots beat photos: zoom in so the table fills the screen before capturing it.
Photograph paper flat, from directly above, in daylight or without flash glare.
Paste cells from another spreadsheet into the grid to combine data from several images.
Only need the text, not a table? Use Image to Text. Working with a text-based PDF? Try PDF to Excel.
Frequently asked questions
How do I convert a picture of a table to Excel?
Load the image, choose the text language and press Extract table. The tool reads every word with OCR, groups words into rows by their vertical position and into columns by the gaps or grid lines between them, and fills an editable grid. Correct anything that was misread, then press Download Excel (.xlsx).
How accurate is the OCR?
Clear screenshots and straight, well-lit photos of printed tables usually come out nearly perfect. Accuracy drops with blur, glare, tilted or curved pages, very small text and handwriting, which OCR engines read poorly. Always check numbers before you rely on them, especially totals.
The columns are wrong. How do I fix that?
Set Columns to the exact number of columns in the table and press Extract table again; the tool then splits the rows at the widest gaps. Cropping the image to just the table, or turning grid lines on or off, also helps. Small fixes are quickest in the grid itself.
Will numbers work in Excel formulas?
Yes. With Store numbers as numbers turned on, cells such as 1250, 3.75 or 12,400 are saved as numeric values, so SUM and other formulas work. Values with currency signs or percent symbols are kept as text so nothing is changed silently.
Can I convert a PDF table?
This tool reads images. For a PDF that contains real text use PDF to Excel. For a scanned PDF, take a screenshot of the page or table and drop it here.
Is my image uploaded?
No. OCR runs in your browser using the open-source Tesseract engine compiled to WebAssembly. Only the engine and language files are downloaded; your image and data never leave your device.
Are all the tools on FreeToolCabin tested for quality and accuracy?
Yes! Every tool on our site undergoes automated Q/A testing in real browser environments to ensure high precision, correct output, and zero errors. If you happen to spot a bug or have a suggestion, please contact us at [email protected] and our team will review it right away.