Home
Sepid eKYC remote authentication system

Sepid eKYC remote authentication system

What is eKYC remote authentication?

Nowadays, remote authentication (eKYC) in order to ensure the recognition of people's identity, is needed to reduce traffic and in-person visits of people to the branches and offices of various organizations and businesses, including state institutions such as government counter offices, the judiciary (especially judicial offices), police offices (especially police+10 offices), municipality (especially urban service offices) and large businesses such as banks, mobile network operators (representative offices), insurance offices, platform and online businesses (such as home services , internet taxi and...). To accomplish this, you can use modern methods based on artificial intelligence processes, encryption algorithms and also receive online inquiries from authorized authorities.

The Sepid eKYC software system is a native product of Sepid System knowledge-based company, which as a part of this company's online authentication solution, and by using various modern methods in the field of biometric authentication and security, can provide remote authentication service in both Cloud and On-premise ways.

This software consists of different sections, which briefly include the following:

Artificial intelligence algorithms

These algorithms include image/video processing and speech processing for face detection, liveness detection, gender detection, age estimation, signature matching, etc.

Security and encryption algorithms

To generate one-time password, digital signature, information encoding, etc.

Inquiries from external authorities

Such as civil registration, Shahkar system, judiciary, educational certificate, police certificate, address and postal code, etc.

Administrative counters

To display performance reports of the remote authentication system

One of the most important parts of the remote authentication solution is its smart modules, which use machine learning algorithms and artificial intelligence to help the system achieve the goals of identity recognition and identification of people. Modules in this field include the following:

1- Face Detection: This module is responsible for detecting the location of the face in the images/videos received from the camera and is able to process the images/videos and detect the location of the face on the server and client side.

2- Face Recognition: The Sepid System face recognition module is one of the most efficient face recognition algorithms which by using the latest deep learning methods, receives a face from an image file (photo) or a playing video (video stream), and after detecting the location of the face, normalizing the image and removing noise, matches the face with the reference photo.

3- Liveness Detection: Vulnerability of facial recognition systems against non-living face display attacks (known as direct attacks or fraud attacks) is one of the main concerns of using this biometric method. The purpose of the presentation attack is to circumvent the face recognition system by displaying an artificial face. Common uses of the fake face include photo printing, electronic display of a face photo, video playback using an electronic display, and 3D face masks. However, to deal with this risk, different methods of detecting liveness in images/videos are used, which include a non-interactive method (Passive) and several active methods (Active) based on challenge-response. Active methods supported in the Sepid eKYC include blink detection, lip reading and speech recognition.

4- Signature Verification, which is used to match two different signatures and can be used in situations where the client's signature sample already exists in the system.

5- Voice Verification, which can be used like the signature matching module for situations where a person's voice sample was taken at the time of registration.

6- Locating the national ID card (NID Detection) and extracting its information with an optical character reader (OCR): This module is responsible for analyzing the structure of the received images and identifying the location of the national ID card and extracting the information written on them. Also, this module is used to convert the MRZ of the passport into personal identity information.

7- Gender Recognition

8- Age estimation

On of the main features of this system is defining a human operator to monitor or answer and make decisions in necessary cases, and if necessary (for example, if the system's approval percentage is low), talk to the client online and via video. Because of the sensitivity of some organizations on security matters and for some services, if necessary, and in order to monitor the correctness of doing business, an observer can be with the client from the start, similar to a two-way conversation.

Among the other capabilities of the system is making inquiries from authorized authorities such as civil registration (receiving identity information and face photo), matching a person's national ID card with mobile number and bank information, inquiring about educational certificates, address (with zip code) and other similar items if relevant access is provided.

The security capabilities of the software include the ability to digitally sign documents, encrypt information, especially stored identity information, and also read passports with NFC to extract people's identity information (and use it as an identity reference).

The management capabilities of the system include its panel and dashboard to provide comprehensive information and reports of service performance on all types of modules, user access management, online monitoring of the system and health checking and ensuring that the services are working.

The micro service architecture of the system and its scalability to respond to the high number of requests are among its other technical capabilities.