Skip to content

KYC Document OCR API

Popular

Read the printed fields from supported Indian identity documents and get them back as structured data. Customers type less, and the checks that follow start from clean input.

Overview

Most onboarding forms in India ask a customer to type details they are already holding in their hand. The result is predictable: transposed digits, names entered in a different order, and dates in the wrong format. Checks then fail for reasons that have nothing to do with the customer. KYC Document OCR reads a supported identity document image, typically Aadhaar, PAN, Voter ID, driving license or passport. It returns the printed values as structured fields your form can pre-fill. Readable fields vary by document type, by image quality and by your service contract. OCR only reads what is printed. It does not confirm that a document is genuine, or that a matching record exists at source. Use it with the verification services, not in place of them: read the document first, then verify the number.

Category
KYC Services
Coverage
India. Supported documents and readable fields depend on your service contract.
Input
An image of a supported identity document, captured or uploaded inside your own flow
Output
The printed values as structured named fields, varying by document type and image quality
Typical users
Lenders and NBFCs, insurers, fintech and payment providers, marketplaces, logistics and staffing platforms, healthcare and telecom
Coverage note
India. Supported documents and readable fields depend on image quality and your service contract.

The overview and the sections after the reference are the FinSour API catalog’s own record for this service, rendered unaltered.

Reference in progress

The request and response reference for this endpoint is not published yet.

Everything below this block is the live catalog record for this service, and it is accurate. The platform’s published API contract does not include this service yet, so there is no request or response detail to document. Rather than fill these sections with a plausible example, the page lists what is missing:

Endpoint URL and method
The host, path and HTTP verb for this service.
Authentication
How a request identifies itself, and where credentials come from.
Request parameters
Field names, types, which are required, and validation rules.
Response schema
The success body, field by field, with a real example.
Error codes
What each failure means and which are worth retrying.
Rate limits
Request ceilings, and what happens at the ceiling.

No sample payload, URL, response body, error code, rate limit or latency figure appears anywhere on this site unless it has been supplied and attributed. Each item in this list is removed only when its own detail arrives; none of them is held back to wait for the rest, and none is ever placed beside a guess in the meantime.

Use cases

  • Lending and NBFCs: pre-fill a loan application from the borrower's uploaded documents
  • Insurance: capture proposal and claim documents photographed in the field
  • Fintech and payment providers: read signatory documents in merchant onboarding
  • Marketplaces and gig platforms: shorten the mobile sign-up form for sellers and workers
  • Logistics and fleet: read a driver's license image before the verification check
  • Healthcare and telecom: capture patient and subscriber identity documents at intake

What this service does

  • Structured fields from a document photograph

    Reads a supported identity document image and returns the printed values as named fields your form can pre-fill, instead of an image someone has to read and retype. What can be read depends on the document type, the layout of the card, the image quality and your service contract.

  • OCR pairs with a verification check

    OCR reports what is printed on the image. It does not establish that a document is genuine or that a matching record exists at source. Extract first to get clean input, then pass the confirmed identifier to the matching verification service for corroboration.

  • Fewer mistyped fields, fewer failed checks

    Manual entry from identity documents is where a large share of onboarding errors begin, and each one resurfaces as a failed check or a support ticket. Pre-filling from the document removes the error before it is made.

  • Made for phone photos, glare and all

    Documents arrive as photographs taken in poor light, at an angle, through lamination glare. Capture guidance in your own screen and a customer confirmation step on the extracted values do more for outcomes than anything downstream of the upload.

How a check runs

  1. Capture the document with consent

    Collect the customer's informed consent, and state which documents you need and why. Capture the image in your own flow, with a framing guide and an on-device check for blur and glare.

  2. Send the image for reading

    Submit the document image through the API for the document types your service contract covers.

  3. Get the printed values back as fields

    Your application receives the printed values as named fields. The fields you get differ by document type, by card layout and by image quality.

  4. Show the values back to the customer

    Display the extracted values for confirmation before you commit them. The customer is the one person who can spot a misread name or date straight away, and they are still in the flow.

  5. Verify the number you extracted

    Pass the confirmed number to the matching verification service, whether PAN, driving license, passport or another supported check. The document is then corroborated against an available record.

Questions

What is a KYC document OCR API?
An interface that reads a supported identity document image and returns the printed values as structured fields. Your application can then pre-fill a form instead of asking a customer or an operations team to retype what is already on the card.
Does OCR prove a document is genuine?
No. A convincing forgery and a genuine card both return clean text, because extraction reports what is printed and nothing about where the document came from. Use OCR alongside the verification checks, passing the confirmed identifier to the verification service for that document type.
Which documents can KYC OCR read?
Supported Indian identity documents, which typically include Aadhaar, PAN, Voter ID, driving license and passport. The document types available to you depend on your service contract, and the fields returned vary by document type, card layout and image quality.
Why do some uploads come back with fields missing?
Usually because of the photograph. Glare across a laminated card, shadows, cropped edges, steep angles and low-resolution images of photocopies all reduce what can be read. A framing guide and a blur-and-glare check in your capture screen improve results more than any change downstream of the upload.
Should I run OCR before or after verification?
Read first, then verify. Extraction pre-fills the form with values the customer only has to confirm, which removes the typing errors that make genuine documents fail. The confirmed identifier then goes to the matching verification service, which reports whether an available record corroborates it.
What consent and data rules apply to document images?
An identity document image is personal data and usually carries more of it than your process needs. Collect informed consent before capture, accept only the documents you require, and keep the extracted fields you actually use. Redact identity numbers that should not be stored in full, restrict access, keep images no longer than the purpose requires, and keep an audit trail.

Catalog entry last updated .