Every day millions of people manually retype information from screenshots, photos, and scanned documents. Whether it's a table from a WhatsApp screenshot, numbers from a photographed utility bill, or text from a scanned form — retyping by hand is slow, tedious, and prone to errors. OCR (Optical Character Recognition) technology makes this unnecessary. Here's how to use it for free.
Step-by-step: convert any image to text for free
- Go to the Screenshot to Document tool at toolbase.work/pages/screenshot-to-document.html
- Upload your image — click Upload or drag and drop your image file (PNG, JPG, WebP supported)
- Select document type — choose Table/Spreadsheet for data grids, Receipt for purchase receipts, or Auto for general text extraction
- Click Extract — the tool reads the text from your image using OCR technology
- Review the result — check the extracted text against the original image for any errors
- Download in your preferred format — CSV, JSON, or plain text
The entire process takes under 30 seconds for a clear, well-lit image. Everything runs in your browser — no upload to a cloud server, no account needed.
What types of images work best?
| Image type | OCR accuracy | Tips for best results |
|---|---|---|
| Desktop screenshot | Excellent | Always prefer screenshots over photos of screens |
| Phone screenshot | Excellent | Use native screenshot feature (side button + volume) |
| Scanned document (good scanner) | Very good | 300 DPI or higher, black text on white background |
| Photo of printed document | Good | Even lighting, camera parallel to page, no shadows |
| Photo of screen | Fair | Avoid — use screenshot instead if possible |
| Handwritten text | Variable | Print clearly in block capitals for best results |
| Low-resolution image | Poor | Increase resolution or retake at closer range |
Common use cases — image to text in real life
Extract data from WhatsApp screenshots
Someone sends you a table, price list, or schedule as a WhatsApp screenshot. Instead of retyping all the data, take a screenshot on your phone, transfer it to your computer (or use the mobile version of the tool), and extract the data directly as CSV or plain text.
Digitize receipts and invoices
Photographing receipts for expense tracking is common, but accounting software needs the data in text form, not image form. OCR extracts the vendor name, date, line items, and total from a receipt photo, ready to paste into your expense system.
Extract marks from result sheets
School and university result sheets often arrive as scanned PDFs or photographed pages. Extracting student names, subject marks, and percentages into a spreadsheet manually takes hours. OCR can process a full page of tabular marks data in seconds.
Copy text from locked PDFs
Some PDF files have copying disabled. Taking a screenshot of the PDF page and running OCR on it extracts the text without needing any special PDF software or permissions.
Extract data from Excel screenshots
A colleague screenshots a spreadsheet and sends it. You need the actual numbers. OCR can reconstruct the table data as a proper CSV that opens correctly in Excel or Google Sheets.
Transcribe text from books or textbooks
Photographing a textbook page and extracting the text is much faster than typing it. This is particularly useful for students who need to quote from physical books that aren't available in digital form.
Why accuracy varies — and how to improve it
OCR accuracy depends almost entirely on image quality. The three biggest factors:
1. Resolution
Higher resolution means more pixel information per character, making characters easier to distinguish. For scanned documents, 300 DPI (dots per inch) is the minimum recommended; 600 DPI gives noticeably better results for small text or complex fonts.
2. Contrast
Black text on white background is ideal. Low contrast (grey text on white, or text on a coloured background) significantly reduces accuracy. Increasing the contrast of your image in a photo editor before running OCR can dramatically improve results for difficult images.
3. Image angle
OCR works best on text that is horizontal and parallel to the image frame. Text photographed at an angle appears distorted — characters that are actually rectangular appear trapezoidal, making recognition harder. Most modern OCR tools include some auto-correction for mild skew, but severe angles still cause errors.
Characters that OCR commonly confuses
Even good OCR makes predictable mistakes. Always check your extracted text for these common confusion pairs:
- 0 vs O (zero vs letter O) — especially in numerical data
- 1 vs l vs I (one, lowercase L, uppercase i) — in some fonts these are identical
- 5 vs S — in poor-quality images or stylized fonts
- 8 vs B — in compressed or low-resolution images
- rn vs m — "rn" can be misread as "m" when characters are close together
- 6 vs G — in certain handwriting styles
For financial data (receipts, invoices, marks), always cross-check totals after extraction to catch any digit errors before using the data.
Extract text from any image — free
Upload a screenshot, photo or scan and get editable text, CSV or JSON instantly. Runs in your browser, no sign-up.
Image to text vs document scanning apps — what's the difference?
Mobile scanning apps like Adobe Scan, Microsoft Lens, and CamScanner combine image capture, OCR, and PDF creation in one flow. They're convenient for on-the-go document scanning. Browser-based OCR tools like ToolBase's Screenshot to Document are better when:
- You already have the image file on your computer
- You need the output as structured data (CSV, JSON) rather than a PDF
- You want structured table data, not just plain text
- You prefer not to install an app or create an account
For the specific task of extracting tabular data from a screenshot to use in a spreadsheet, a browser-based tool with CSV export is the most direct workflow.
Privacy: what happens to your image?
ToolBase's Screenshot to Document tool processes images using Tesseract.js — an OCR library that runs entirely within your browser using JavaScript. Your image is never uploaded to any server and never leaves your device. Once you close the browser tab, no record of the image or extracted text is retained anywhere. This makes it suitable for processing sensitive documents like financial statements, medical records, and confidential business data.