Photo table to Excel
Read a regular table photo cell by cell, correct the text, and export XLSX or CSV. Korean and English OCR runs in your browser.
Photo and recognition settings
Edit the public example or choose a photo. The example results can be exported before recognition.
Review cells and save
A new photo is selected. The results below belong to the previous input. Read the new photo before saving.
The image is divided using your equal grid spacing. Correct empty or incorrect cells. XLSX stores all cells as text. CSV protects formula prefixes with an apostrophe.
How it works and limits
The public example is a 3-row by 3-column table with item, quantity and price. Keep rows and columns at 3 and 3 respectively and press Read photo to run actual OCR on nine cells. Check Pencil, quantity 3 and price 1500 against the image, edit mistakes and save XLSX. Crop the image closely to the table boundary; row heights and column widths must be equal, with one text line per cell. JPG, PNG and WEBP are limited to 12 MB, 6,000 pixels per side and 12 million pixels, then reduced to at most 2,200 pixels / 4 million pixels for recognition. Up to 48 cells are supported. Merged cells, skewed tables, handwriting and Japanese recognition are unsupported.
Frequently asked questions
Does this detect the table structure automatically?
No. It divides the photo evenly using the row and column counts you enter, then reads each cell. Crop irregular tables into smaller pieces or correct cells manually.
Are numbers and formulas preserved as Excel values?
All XLSX cells are text, preserving leading 0s and formula-like strings. Convert reviewed values to numbers in Excel if you need calculations. CSV formula prefixes are protected with an apostrophe, which can change the displayed text.
Are photos or results sent to a server?
Photos and results are processed only in this browser. OCR code and Korean/English language files download from the same site. Reloading clears your results.
Calculation criteria and reference materials
- Tesseract.js — configure browser OCR API Worker
API in the browser Worker that recognizes images and fixes the worker·core·language file path to the same site. Accuracy of recognition results is not guaranteed.
Verification of calculation criteria: · Calculation·verification principles · Report errors