/Veridas digital onboarding, first to meet Spanish’ National Cryptologic Center (CCN) new requirements for issuance of qualified certificates

CERTIFICATION AND EVALUATION REQUIREMENTS

In this section, where reference is made to the required certifications and evaluations, it is worth mentioning how the CCN demands all qualified solutions to have their biometric engines evaluated by the National Institute of Standards and Technology, NIST, considered the most prestigious independent evaluator body in the world. Veridas has always submitted its facial and voice biometric engines for evaluation by the NIST, obtaining fantastic results yearly. 

Specifically, the guide requests that “the biometric facial matching system between the applicant and the ID photo must have been evaluated, according to the Face Recognition Vendor Test (FRVT) in the VISABORDER category, by NIST and have obtained an FNR (False Negative Rate) of less than or equal to 5% for a FPR (False Positive Rate) of less than or equal to 1/1,000,000. The database used for the test must be the one used by NIST in 2020 or higher” (Article 24).

In this regard, Veridas obtained an FNR rate of 0.80% in its last submission in September 2021, well below the rate required by the CCN, which places it among the best biometric engines worldwide.

FUNDAMENTAL SECURITY REQUIREMENTS (RFS)

1. Protection against evidence capture attacks: the tool must ensure that a single device is used in a single sequential act of time. In addition, it must ensure that the process is executed in real-time without allowing pre-recorded files. 
2. Biometric verification: the tool should provide facial biometric verification through the captured selfie and document. It should also incorporate active or passive liveness detection and implement presentation attack detection (PAD) mechanisms.
   To assess the biometric verification requirements, the CCN specifies that the evaluator should do this by applying IT-14, also known as the biometric evaluation module (MEB). These requirements are explained later in the article. 
3. Auditing: the annex defines different characteristics that the auditing tool accompanying the manufacturer’s solution must have. 
4. Secure communications: the tool must establish secure channels when exchanging sensitive information with authorized entities or between different parts of the product, following the guide CCN-STIC-807.
5. Trusted administration: the tool shall ensure that only a user with administrator permissions can perform administrative functions, distinguishing between the role of configurator and auditor. 
6. Identification and authentication: the tool shall identify and authenticate each user before granting access, implementing mechanisms to prevent attacks, protecting the confidentiality and integrity of authentication credentials, and blocking or close the session after a certain period of inactivity.
7. Protection of credentials and sensitive data: if the tool stores credentials and/or other sensitive data, these should be stored using the cryptological mechanisms authorized in CCN-STIC-807.

    

VALIDATION OF SUBMITTED DOCUMENTS

This section of Annex F.11 details the following requirements regarding the validation of submitted identity documents: 

1. The tool shall implement detection mechanisms for replay attacks and print attacks. 
2. The tool shall be able to verify that the validity date of the document has not expired.
3. The tool shall check the data integrity of the Visual Inspection Zone (VIZ) with the Machine Readable Zone (MRZ).
4. The tool shall generate alerts to the operator whenever it detects any of the described attacks or failed test checks.

BIOMETRIC EVALUATION MODULE (MEB)

As mentioned above, the MEB establishes a set of tests to be performed by the laboratory on the software under evaluation (Target Of Evaluation or TOE) to verify the biometric verification capability of the tool under assessment. In this set of tests, six types of attacks to be prevented are distinguished: 

1. Imposter presentation attacks: this test aims to verify that the TOE is not vulnerable to the impersonation of a bona fide user by an attacker resembling the bona fide user.
2. Presentation attacks using videos: this test aims to check that the TOE is not vulnerable to the impersonation of a bona fide user by an attacker using a video of a bona fide user. 
3. Presentation attacks using very low-cost masks: this test aims to verify that the TOE is not vulnerable to an attacker’s impersonation of a bona fide user using a low-cost (e.g., paper) mask as an artifact. 
4. Presentation attacks using advanced masks: the objective of this test is to verify that the TOE is not vulnerable to the impersonation of a bona fide user by an attacker using an advanced mask as an artifact. 
5. Presentation attacks using makeup: this test aims to verify that the TOE is not vulnerable to the impersonation of a bona fide user by an attacker using makeup as an artifact.
6. Attacks using deepfake computing tools: this test aims to verify that the TOE is not vulnerable to an attacker’s impersonation of a bona fide user using deepfake tools. These tools can manipulate video and sound, superimposing such those elements on the actual images. Focusing on a possible use of identity theft, deepfakes would be used to modify the face of one person on the video where another person appears.