FileKitHow to OCR Scanned Documents — Extract Text from Images
A guide to Optical Character Recognition: what it is, how to use browser-based OCR, tips for better accuracy, and when to use OCR vs. direct text extraction.
What Is OCR?
OCR (Optical Character Recognition) converts images of text into actual, selectable, searchable text. If you have a scanned contract, a photographed whiteboard, or a screenshot of a conversation, OCR extracts the text so you can copy, edit, search, and process it.
How to OCR a Document
1. Browser-Based OCR
FileKit's OCR tool uses Tesseract.js — an open-source OCR engine compiled to WebAssembly — to recognize text entirely in your browser. Supports English, Simplified Chinese, Japanese, and mixed English+Chinese. Drop an image or scanned PDF, choose the language, and get the extracted text in seconds.
2. Google Drive
Upload a scanned PDF or image to Google Drive, right-click → Open with → Google Docs. Google applies OCR automatically and creates an editable document. Works well for simple layouts but struggles with complex formatting.
3. Adobe Acrobat
Acrobat's "Scan & OCR" feature creates a searchable text layer on top of scanned pages. The original image stays intact while text becomes selectable and searchable. Best for archival quality.
Tips for Better OCR Accuracy
- Resolution matters. Aim for 150 DPI minimum, 300 DPI for best results. Low-resolution phone photos of documents often produce poor accuracy.
- Contrast is key. Dark text on a white background gives the best results. Colored backgrounds, watermarks, and low-contrast text degrade accuracy.
- Straighten the image. Skewed or rotated text confuses OCR engines. Align the document before scanning or photographing.
- Choose the right language. Always select the primary language of the document. Mixed-language documents work best with the combined mode (e.g., English + Chinese).
OCR vs. Text Extraction
Not all PDFs need OCR. If your PDF was created digitally (exported from Word, Chrome's "Print to PDF", etc.), the text is already embedded and selectable. Use the PDF to Text tool instead — it is faster and more accurate because it reads the text directly rather than guessing from pixel patterns.