Capturing data from identity documents and verifying someone’s identity are necessary procedures for companies in many industries. For financial institutions and banks, for example, these procedures are part of the client onboarding process and required by strict KYC regulations.
Doing this manually is not the most efficient option. It is not scalable (only by hiring more staff) and even then, human errors such as typos can hurt you more than they benefit you.
In any of these cases, using an Intelligent OCR solution is a better alternative to automatically extract data from ID documents and verify someone’s identity with it.
If you want to know more about the smartest way to process passports and ID cards, you are at the right place. Let’s have a look at the use of automated solutions to speed up ID-related business processes!
Smart vs. traditional client onboarding
No matter what type of business you run, it is necessary to streamline the onboarding process of new clients in the fastest, most practical and efficient way. It is the key to increase the conversion rate and improve the user experience. But how can you achieve this?
Let’s say a customer wants to open a bank account. Traditionally, they would have to go to the bank and show their ID. The bank clerk would manually copy the information from the ID and they would be confirmed as the person opening the account. This was customer onboarding back in the days. There must be a smarter way to complete this process.
What is the smart solution?
Nowadays, there are smarter ways to capture data from ID cards and passports and verify someone’s identity. With Optical Character Recognition (OCR) and AI, for example, all sorts of documents containing text can be read and data can be extracted from it automatically.
The software takes over the manual process of reading and checking documents. Any kind of document containing information about the potential customer can be run through it. The software automatically recognizes the text that it needs and makes it extractable.
The extracted data can then be used according to your needs, such as passing it on to any desired system, database or tool for further processing.
What does the OCR process look like?
Now that we know about automated document processing solutions, let’s have a look at how these work on ID cards and passports.
The process typically consist of 4 steps:
- Document entry
- Text recognition
- Parse to structured data
- Signature and Photo extraction
1. Document entry
The first step of the process is to provide a picture or PDF file of an ID document to the software. Nowadays, most OCR software can be used on smartphones. With Klippa’s OCR software for ID cards, for example, you can upload a picture by simply taking a photo with your camera. But it’s also possible to send it in through the web application or e-mail.
The software will automatically crop the image of the document and optimize the image quality.
2. Text recognition
As soon as the image is received, the OCR software analyzes the patterns of light and dark that make up the letters and numbers to transform the scanned image into text. All the data you need is extracted by the OCR software in a matter of seconds.
In this step also duplicates, errors and fraud can be recognized automatically. Forged images of documents, for example, can be identified with Photoshop detection and exif analysis.
All text from the ID document is extracted, but it’s not yet structured.
3. Parse to structured data
The text, which was gained from the previous step, is converted into a structured JSON file or an alternative of your choice, ready for you to process.
From this point on, the information from the ID document can easily be processed in the database of your company. One last option is to extract the signature or the picture from the ID document for the verification process.
4. Signature and Photo extraction
To improve the security of the onboarding process, it is possible to identify, extract, crop and convert any picture and signature on an ID document into a usable format.
This works as follows. The software uses computer vision and deep learning to analyze the document section by section, dividing it into text and visual components.
After that, it checks what visual components are likely to be a signature or photo. If it finds a signature, the software crops the signature, removes background patterns and extracts the leftover scribble into a usable format.
In the same way, photos can be cropped and extracted from the document. The signature and photo can then be used for identity verification purposes through signature and selfie comparison.
Klippa’s data extraction and identity verification solutions are used in many industries, usually for KYC and onboarding processes.
Let’s have a closer look at two cases that we come across most often:
- Digital onboarding with ID documents and e-signature
- Identity verification for KYC check with selfie biometric match
Digital onboarding with ID documents and e-signature
In order to ensure a stable and compliant service, many companies require a person’s ID document at sign up. Technology and software can make a big impact here.
Take the case of vehicle rental services, for example.
When it comes to renting a car, customers usually don’t want to get inside an office to fill in forms and contracts. The customer wants to pick up the vehicle and leave as soon as possible. ID data extraction and e-signature makes the onboarding process and signing of contracts smooth and efficient.
Car rental companies can perform the customer onboarding from mobile devices or the web. Intuitively and quickly:
- Upload a photo of the ID document to the software
- Extract all the data and customer only needs to confirm it
- E-sign for contracts
- Verify identity with a signature match
Customers can be uploaded into the database remotely even before picking up the vehicle.
Identity verification with selfie biometric match
Banks must make sure that their clients are genuinely who they claim to be. They are bound to KYC procedures, which are mandatory to verify a client’s identity when opening an account.
KYC procedures defined by banks involve all the necessary actions to ensure their customers are real, and that they assess and monitor risks properly. These processes help prevent and identify money laundering, terrorism financing, and other illegal financial schemes.
Digital ID verification helps banks provide a smooth customer onboarding experience that complies with KYC regulations and minimizes the risk of frauds.
The verification process could look like this:
- Scan or upload ID document
- Extract information from ID document
- Verify identity with selfie biometric and signature check
In such a digital and automated process, a customer’s identity can be verified almost instantly. The following video summarizes the steps in this process:
Klippa’s solutions for other identity documents
What if you want to extract data from another identity document? Besides passports and identity cards, Klippa is trained for any type of identity document. We support:
- driving licenses,
- residence permits,
- student IDs
- and more.
We got you covered!
How can Klippa help in automating your business processes?
Klippa is the solution for extracting data and verifying someone’s identity with ID cards and passports. Our software is enhanced with technologies such as OCR, machine learning and computer vision, which empowers all the features that you need for KYC checks and customer onboarding.
Klippa complies with GDPR regulations and responds to the demands of users and companies.
We would gladly hear about your specific case and talk about how we can help to speed up your organization.
If you want to know more about our solutions, plan a demo below or get in touch with us. We would be happy to help!