Understanding OCR Results
When you upload a document to DocSnap, it's processed by an OCR (Optical Character Recognition) engine that extracts text from the image. This page explains how OCR works and what to expect.
How OCR Processing Works
- Upload -- your document image is saved securely on the server
- Queue -- the document enters a processing queue
- Processing -- the OCR engine analyzes the image and extracts text
- Completion -- extracted text is stored and indexed for search
Processing typically takes a few seconds per document. The app polls automatically and updates the view when results are ready.
What Gets Extracted
The OCR engine reads all visible text in the document, including:
- Headers and titles
- Body text and paragraphs
- Table contents (row by row)
- Reference numbers and codes
- Dates, addresses, and contact information
- Fine print and footnotes
OCR Result Display
Results appear in the document detail view with:
- Engine -- the OCR technology used (shown as a badge)
- Language -- the detected language of the text
- Extracted text -- displayed in a monospaced font for easy reading
The text is fully selectable -- long-press to select, then copy to your clipboard.
Common Document Types
DocSnap is designed for freight industry documents. Common types include:
| Document Type | Typical Content |
|---|---|
| Bill of Lading (BOL) | Shipper, consignee, carrier details, shipment items, weights, reference numbers |
| Proof of Delivery (POD) | Delivery confirmation, receiver name, date/time, condition notes |
| Freight Bill | Charges, line items, invoice numbers, payment terms |
Improving OCR Accuracy
If OCR results aren't as accurate as expected, try these tips:
- Better photos -- ensure good lighting and focus (camera tips)
- Higher resolution -- closer photos capture more text detail
- Clean documents -- remove wrinkles, shadows, and obstructions
- Correct orientation -- photograph documents right-side up
Editing Results
If the OCR engine misreads any text, you can manually correct it. See Editing OCR Text for details.
Searchable Text
All extracted text is automatically indexed for full-text search. This means you can find any document by searching for words that appear in it -- reference numbers, names, addresses, or any other text. See Full-Text Search for details.