Extract Table from Image
Extract a table from a photo or screenshot into CSV using OCR. Runs entirely in your browser.
Extract a table from an image
Upload a photo or screenshot of a table, choose a language, then run OCR to reconstruct it as CSV.
Extracted table
Note: table reconstruction is a best-effort heuristic based on word positions, not a true table-structure parser. It works well on simple grid-like tables with clear row and column alignment, and less well on complex tables with merged cells.
How to use this tool
- Drag and drop a photo or screenshot of a table into the box above, or click to browse and select one.
- Choose the language that matches the text in your table.
- Press Extract table and wait a moment while OCR runs and the table is reconstructed.
- Check the preview table, then download the result as a CSV file.
FAQ
Frequently asked questions
Is my image uploaded anywhere?
No. Everything runs entirely on your device using Tesseract.js, an open-source OCR engine compiled to WebAssembly. Your image is never uploaded to Calcooo or any server.
How accurate is the table structure?
This is a best-effort reconstruction based on the horizontal and vertical position of recognized words. It works well on simple, clearly-aligned tables with consistent spacing, but may misalign columns on tables with merged cells, rotated text, or very tight spacing.
What format do I get?
A CSV file you can open directly in Excel, Google Sheets, or any spreadsheet app, plus a live preview table on this page so you can check the result before downloading.
Can I fix mistakes before downloading?
Not directly on this page — download the CSV and correct any misaligned cells in your spreadsheet app, or retake the photo straighter and with more even lighting for a better result.
