Passport reading: Datasets & Models
A computer vision workflow that locates and reads the machine-readable zone (MRZ) on passport pages, extracting identity fields from the passport document image. It combines MRZ region detection with OCR to turn a photographed or scanned passport into structured text.
Passport reading
This workflow performs passport MRZ detection by locating the machine-readable zone on scanned or photographed passport pages and running OCR to decode its character lines into structured identity fields. It takes passport document images as input and detects the MRZ region along with supporting areas like the photo and printed data fields to support cropping and accurate recognition.
Try it out
Powered by general detection model
Run on custom image
Drop an image here or click to upload
Why automate passport reading
Manually keying passport details at check-in desks, KYC onboarding, or border control is slow and error-prone, and a single transposed digit in a document number can break downstream verification. Automating passport MRZ detection lets you capture the biographic page once and reliably parse the standardized MRZ format, using its built-in check digits to validate the read. For Document AI and OCR teams, this removes a tedious data-entry bottleneck and reduces the review workload for identity documents. It also enables straight-through processing where clean captures never need a human to retype anything.
Key benefits
Teams get faster, more consistent identity capture with fewer transcription errors, and MRZ check-digit validation gives an immediate confidence signal on each read. The workflow runs on the cameras, scanners, and mobile devices you already deploy, so it slots into existing onboarding, travel, and hospitality intake flows. Because detection and OCR models are trainable on your own document samples, accuracy scales as you add more passport variants and capture conditions.
How to use it
- Upload an image: Drop any photo into the widget above to try passport reading for free, right on this page.
- Preview instant results: The model detects passport, machine-readable zone, passport photo, printed text field, document edge within seconds and returns an annotated image with structured predictions, so you can immediately see how well it works. Adjust the detection classes and re-run until the results fit your scenario.
- Like what you see?: With a free Roboflow account, you can deploy this workflow with the Serverless Cloud API, customize it in your workspace, or train OCR / document model on your own images for production accuracy.
- Get started for free: No code or training data required to begin: upload, review the results, and make it part of your stack in a single click.
Recommended model for passport reading
When you're ready to build a production model for passport reading, we recommend training OCR / document model. Unlike the ready-to-use model in the demo above, you fine-tune it on your own labeled images for higher accuracy, then deploy to a Serverless Cloud API endpoint.
Not sure where to start?
Describe your Passport Reading project to Roboflow Agent.