How to convert image to CSV
- Take a clear snapshot or screenshot of your printed table and add the picture to the converter.
- Choose the matching language under "Language of the text" to aid accurate character recognition.
- Select your required delimiter using "Separator" and check the UTF-8 BOM option for Windows Excel if needed.
- Convert the image to run local OCR and save your structured CSV file ready for database or spreadsheet import.
Paper price lists, timetable digitization, and private in-browser OCR
Retyping documents such as classroom rosters, supplier price lists, or utility statements consumes valuable office hours and introduces mistakes. This utility reads tabular graphics right inside your browser, allowing desk workers to convert scanned records or smartphone photos into clean CSV spreadsheets without uploading files to cloud services. The resulting UTF-8 encoded .csv file opens smoothly in Excel, Google Sheets, LibreOffice Calc, Numbers, and office databases.
The reconstruction algorithm groups recognized words into rows based on baseline coordinates. Horizontal gaps wider than roughly one letter height create new cells, while cells aligning vertically across rows are matched into common columns, accommodating right-aligned sums. Tilted desk photos are straightened using the slope of text lines, and isolated title rows stay anchored in column A.
Separator settings, character encoding, and scanning guidelines
Office workflows often require specific delimiter conventions, which you can choose using "Separator", featuring "Comma ,", "Semicolon ; (European Excel)", and "Tab". Values containing the active delimiter, quotation marks, or line breaks are enclosed in quotes according to RFC 4180 rules. When preparing files for Windows Excel, tick "Add a UTF-8 BOM (Excel on Windows shows accents correctly)" so accented names render correctly. The "Language of the text" setting includes English by default, with additional packs for German, Spanish, French, Italian, Portuguese, Dutch, Indonesian, Polish, Turkish, and Vietnamese available via a single download.
Extracted values are recorded exactly as read without assigned data types, leaving interpretation to your spreadsheet software. Images smaller than 1200 pixels on their long side are automatically doubled in size before reading to clarify characters. Optical recognition extracts text only, omitting borders, background colors, merged cells, and formulas; text wrapped across two lines inside a single cell produces two rows, and handwriting reads poorly. Always verify recognized figures against your original document, because OCR routines can occasionally mistake a zero for an O, a one for an l, or an eight for a B. If no text is detected, a clear notification is displayed. To retain numeric types in a workbook, use Image to Excel, or try PDF to CSV and CSV to Excel.
Questions
Are sensitive student grades or company financial snapshots uploaded to any server?
No. All optical recognition takes place strictly within your web browser, ensuring private office paperwork remains securely on your computer.
Why should I check the UTF-8 BOM checkbox when opening the CSV in Windows Excel?
The UTF-8 BOM tells desktop Microsoft Excel on Windows to interpret special characters and accented letters properly instead of showing corrupted symbols.
Can the converter read handwritten price tags or paper ledger notes?
No. The optical engine is engineered for typeset print and cannot accurately interpret cursive handwriting, pen marks, or manual annotations.