You have a photo of a document on your phone — maybe a receipt, a page from a book, a handwritten note, or a business card. You need the text from that image, but retyping it manually is tedious and error-prone. The solution is OCR: Optical Character Recognition.
What You Will Need
All you need is the image file (JPG, PNG, or WebP) and a web browser. No software to install, no accounts to create. The Scan Image tool handles everything right in your browser.
Step-by-Step Guide
- Take a clear photo. Good lighting, minimal shadows, and a straight angle make a huge difference in OCR accuracy. Hold your phone directly above the document and make sure all text is visible and in focus.
- Open the Scan Image tool in your browser.
- Upload your image(s). You can upload multiple images at once if you have several pages to process.
- Select the language. Choose the language of the text in your image. English is the default, but Spanish, French, German, and Portuguese are also supported.
- Click Extract Text. The OCR engine will analyze your image and extract all readable text.
- Copy or download the text. The extracted text appears in a text box. You can copy it to your clipboard or download it as a .txt file.
Tips for Better OCR Results
Image quality matters most
The single biggest factor in OCR accuracy is image quality. A sharp, well-lit photo with high contrast between text and background will produce much better results than a dark, blurry, or angled shot.
Straighten the document
If the document is at an angle in the photo, the OCR engine has to work harder to read it. Most phone cameras have a document scanning mode that automatically straightens and enhances the image.
Crop out unnecessary areas
If your photo includes a lot of background around the document — the desk, your hands, other objects — crop the image to just the document before uploading. This helps the OCR engine focus on the text.
Use high resolution
Higher resolution images contain more detail, which translates to better OCR accuracy. Do not downscale your images before uploading them.
What OCR Can and Cannot Do
Understanding the limitations of OCR helps you set realistic expectations and get the best results.
OCR works very well with:
- Printed text in standard fonts like Arial, Times New Roman, Calibri, and Helvetica
- Clean, high-contrast documents — black text on white paper produces the best results
- Multiple languages — when the correct language is selected in the tool settings
- Receipts and invoices — the structured format helps OCR identify text regions
OCR struggles with:
- Messy handwriting — printed handwriting works reasonably well, but cursive is much harder to recognize
- Very small text (under 8pt) or text in decorative or script fonts
- Documents with complex backgrounds, patterns, or watermarks that interfere with character detection
- Low-quality or heavily compressed images — JPEG artifacts can confuse the recognition engine
- Tables and forms with dense grids — the engine may mix up data from adjacent cells
Scanning Multiple Pages at Once
If you have several pages to scan — like a multi-page letter or a stack of receipts — you can upload all the images at once. The tool processes each image sequentially and labels the output by image name, so you can easily identify which text came from which page. This batch processing approach saves significant time compared to scanning images one by one.
For best results with multi-page documents, number your image files before uploading (page-01.jpg, page-02.jpg, etc.) so the output appears in the correct order.
Language Support and Accuracy
The OCR engine supports multiple languages, each with its own trained model. Selecting the correct language is critical for accuracy — an English model will not correctly recognize accented characters in French or Spanish, and vice versa. The supported languages include English, Spanish, French, German, and Portuguese.
For documents that contain text in multiple languages, choose the primary language. The engine will still attempt to recognize characters from other languages, but accuracy may be lower for the secondary language.
Building a Scanning Workflow
If you regularly need to digitize paper documents, establishing a consistent workflow saves time:
- Photograph documents in good lighting with your phone camera or a dedicated scanning app
- Crop and straighten images before uploading
- Upload to the Scan Image tool and extract text
- Review and correct any OCR errors in the output
- Save the text in your preferred format — paste into a document, save as .txt, or convert to PDF
Many phone cameras now include a built-in document scanning mode that automatically enhances contrast and corrects perspective. Using this mode before uploading to the OCR tool significantly improves accuracy.
After Extracting the Text
Once you have the text, you can:
- Paste it into a Word document or Google Doc for editing
- Convert it to PDF for sharing
- Use it for data entry or record-keeping
If your source is a scanned PDF rather than an image, use the OCR PDF tool instead — it processes each page of the PDF automatically.
Final Thoughts
Converting images to text used to require expensive software. In 2026, you can do it for free in your browser with impressive accuracy. Whether you are digitizing receipts, extracting text from a book page, or transcribing notes, OCR saves you from the drudgery of manual typing.
For a deeper understanding of the technology behind this, read our explainer on what OCR is and how it works.