Go beyond OCR and verify Passports using Affinda’s Powerful ID Extraction API. Our unique technology sees documents like a human, offering unprecedented accuracy.

Want To Parse Passport Data?


Step 1: User Uploads Passport Scan

You send the passport to our API.


Step 2:  We parse the data quickly

Our powerful AI Engine, VEGA, identifies and extracts the key information with 99%+ accuracy.


Step 3:  You get clean, formatted data

And just like that, the passport is verified.

Accurate & Fast Passport Verification

Using our Passport OCR SDK, you can quickly extract passport data and send it for verification.

You can accept all forms of images from your users – PNG, JPG, PDF, JFIF – you name it, our AI can handle it. Affinda’s AI will process the data and send it to you, automatically. This is all based on our ‘better than human’ AI which is more advanced than traditional OCR techniques and has a dramatically lower error rate.

Passport document

Process These Passport Fields Automatically Using our API

Family name

First name

Middle name



Country of issue (code)

Date of birth


Date of issue

Date of expiry

Place of Birth


Type of passports



Finger print (if available)

MRZ  Type of document

MRZ  Issuing country

MRZ  Family name

MRZ  Given name

MRZ  Passport Number

MRZ  Nationality code

MRZ  Date of birth

MRZ  Sex

MRZ  Expiry date of passport

MRZ  Personal number

What Makes Our Passport OCR API Special?

  • Accurate: World-class accuracy delivered by VEGA, our groundbreaking AI
  • Fast: Most passports processed in less than 1 second
  • Secure: Use our self-hosted solution to process passports securely
  • Flexible: Work with our team to customize the solution to your needs

Vega: Our Intelligent Document Processing Engine

VEGA processes documents combining human-like understanding and computer-grade accuracy combining the latest in the AI advancements

Natural language processing

Computer vision

Deep learning

VEGA logo

Where Affinda really stood out out head and shoulders above the others is in their level of support and attention to the customer.

Scott S
Verified Review

Overall, we are very happy with having switched. Problems are resolved quickly. We have some unique requirements and they were able to work with us on that.

Jon S
Verified Review

AI Experts. Specialists in Document Extraction.

Affinda is a team of AI nerds, headquartered in Melbourne, Australia and with a global team spanning the world. We specialize in solving novel document extraction problems and have built custom AI solutions for use cases across Insurance, Banking, Investment, Logistics, Recruitment and Finance.

Get in Touch to Discuss Your Project

We can customize our technology to meet your specific needs. With teams in Europe, North America and Australia, we can help – no matter where you are.

Frequently asked questions:

What is ID document extracting?

Our technology is used to extract all the important information from an ID document. Modern ID document parsers like Affinda's leverage multiple AI neural networks and data science techniques to extract structured data. Typical fields being extracted may include the document number, first name, middle name, last name, date of birth, address — plus much more. Our ID document extraction tools can be integrated into a software platform, to provide near real time automation.​

Can the Affinda ID Extractor extract information from all ID documents?

Affinda’s ID document OCR extractor is capable of extracting data from Passports, Driver’s Licenses and Birth Certificates. If you require additional ID documents and fields to be extracted, please reach out to our AI experts.​

What artificial intelligence technologies does Affinda use?

Good intelligent document processing requires a combination of technologies and approaches.​​

Our solution uses deep transfer learning in combination with recent open source language models, to segment, section, identify, and extract relevant fields.​​

We use image-based object detection and proprietary algorithms developed over several years to segment and understand the document, to identify correct reading order, and ideal segmentation.​The structural information is then embedded in downstream sequence taggers which perform Named Entity Recognition (NER) to extract key fields.​Each document section is handled by a separate neural network.​Post-processing of fields to clean up location data, phone numbers and more.​Comprehensive skills matching using semantic matching and other data science techniques​

What is OCR, and how does Affinda's ID parser utilise it?

OCR stands for Optical Character Recognition. Affinda’s software uses OCR to convert an image that doesn’t have text data — such as a photo of an ID document — into a file that is machine-readable. ​

How secure is the solution for sensitive documents?

All uploaded information is stored in a secure location and encrypted. You can read all the details here.

What format is the output in?

We export the data from our ID documents in JSON format so that it can be most easily integrated into your platform or database.​

What if I don’t see a field I want to extract?

We’d be happy to build you an ID document parsing tool with custom fields specific to your use case. Just click here to get in touch with our development team!

Can you extract data from other ID documents?

We’d be happy to build you an ID document parsing tool for another document that we don't currently have listed. Click here to get in touch with our development team and we can discuss your use case.

How long does it take to parse IDs?

Our Online App and ID document API will process documents in a matter of seconds. If you’re looking for a faster, integrated solution, simply get in touch with one of our AI experts.

Can I license Affinda's ID parsing functionality for my own platform?

Our ID document reading functionality is designed to be flexible, easy to integrate, and simple enough to white-label. Click here to get in touch with our sales team and discuss licensing within your app or platform.​

How does Affinda's ID parser learn and evolve?

Affinda's ID document extracting software uses machine learning and natural language processing to observe and adapt to differing documents. The software uses a dynamic approach where it is uses human like logic to recognise different document types and characteristics and build and apply a library of logical reasoning

How much does Affinda's professional ID extractor software cost?

Pricing varies depending on the integrations you’ll be using, as well as the number of ID documents you’ll be parsing each month. Please get in touch with our sales team to learn more!​

How can I trial Affinda's ID document parser?

Simply sign up for a free trial​ and start uploading your passports, licenses and birth certificates in bulk, in any common image or text format, such as .pdf, .doc, .docx, .odt, .png or .jpg.  That’s all there is to it!​

Which document formats does Affinda's ID document extraction software accept?

Our free online ID document parser accepts .pdf, .doc, .docx, .odt, .png and .jpg formats.​