The first step is to provide an image or PDF file of the payment card to our API. Usually this is done from a mobile app, email, FTP, or web application.
The document can be sent both cropped (without background) and uncropped (with background). If the image is sent uncropped, our API automatically crops it to the right size. There is also a possibility to use the Klippa scanning SDK in mobile apps.
Step 2 : Image to text with OCR
As soon as an image or PDF with a card number has been submitted, it is converted to a TXT file. In this step, all text from the file type is extracted, but not yet structured.
Step 3: JSON output from the API
The Klippa parser takes the TXT gained in step 2, and converts it into structured JSON by using machine learning. The JSON is returned as output from the API. The card number is now extracted.
Now, the field can easily be processed into your database. Whether you are processing card numbers for age verification or KYC processes, Klippa is here to help you.
The image on the left is a simplified example of the JSON response.
Why use Klippa’s data capture software?
Spend less on extracting card numbers from payment cards by using Klippa.
Process card numbers automatically within seconds.
Prevent manual data entry errors with high quality extraction of card numbers.
Automatically recognize errors, duplicates, and fraud.
Other data fields that Klippa can capture from payment cards
Below is a list of data fields which Klippa can capture and extract from payment cards. This is not an exhaustive list, but depends on what the client requires.
Frequently Asked Questions
What does OCR for card numbers cost?Which types of bank cards are supported by our OCR solution?What are common use cases?Which languages does Klippa support?Is Klippa card number processing GDPR-compliant?Can Klippa convert files to CSV, XLSX, XML or JSON?
Which types of bank cards are supported by our OCR solution?
Klippa can extract data from payment cards from any country. All languages with the Latin alphabet are supported.
Cards from common vendors like MasterCard, Visa, American Express are supported, but also less common card brands can be processed.
What are common use cases?
When it comes to card number OCR, common use cases are age verification, KYC, user onboarding and payments.
Manually typing over card numbers into your database is no longer efficient or reliable. Whatever your use case is, you can replace manual data entry with Klippa.
Which languages does Klippa support?
Klippa supports all European languages. Our engine performs best on file types in English, Dutch, Norwegian, Danish, Swedish, Finnish, Italian, Portuguese, Spanish, German, Hebrew, and French.
Other languages can be supported on request. We are happy to train our machine learning models to assist you.
Is Klippa card number processing GDPR-compliant?
All services that Klippa offers are fully GDPR-compliant. By default, we use ISO-certified servers within the European Union to process documents.
If you are not located within the EU, we can also set up servers in a region of choice.
Next to that, a data processor agreement is always in place. We don’t store any of your or your customer’s data after processing.
Can Klippa convert files to CSV, XLSX, XML or JSON?
Yes we can.
Klippa takes pictures of the payment card, or their PDF equivalents, and converts them to readable text using OCR. From there, we use machine learning to turn text into structured data.
Most of the time, we use JSON, but depending on your preference we can also convert documents to CSV, XLSX or XML.
Do you have any questions about Klippa DocHorizon? Get in touch by mail, phone or chat!
+31 50 2111631