Scan your Document

📄 Step 3.1: Scan Your Document

Once the user is redirected to the IdvWebServer with a valid web_session_id, the identity verification begins with Step 1: Document Scan.

🎯 Purpose

The user uploads or captures an identity document. IdvWebServer automatically detects document type and extracts key fields using OCR (Optical Character Recognition).


🧾 Document Submission Options

The user can submit the identity document through one of the following methods:

  • 📷 Take a Photo
    Opens the device’s camera. The user aligns the document within the frame and captures an image.



  • 💻 Browse from Computer
    Uploads a scanned image or file from the user’s system (JPEG, PNG, PDF).


After uploading the front side, the user can:

  • Confirm to proceed, or
  • 🔁 Retake/Re-upload the image if unclear.

If the document has a back side, the system prompts the user to:

  • 📷 Take Back Photo, or

  • 📁 Upload Back Photo


⚠️

Tip: It's recommended to upload both sides for better verification accuracy.
Users can skip the back side by clicking “Skip and Continue” if it's not mandatory.


⚙️ Backend Behavior

  • 🛡️ Validates image format and size
  • 🔍 Performs OCR and identifies document type
  • ✅ If V3 scan structure is enabled via sdk.app.settings, extra compliance checks are triggered

👨‍💻 Developer Notes

  • 🎥 Ensure the browser allows webcam access
  • 🧠 Backend handles all scanning logic—no enterprise API calls are required at this step
  • 📑 Supports both front-only and front+back document flows