Biometric Self Enrolment Feasibility Trials (BSEFT)
The ground-breaking Biometric Self Enrolment Feasibility Trials (BSEFT) play a pivotal role in the Home Office’s ambition to provide immigration customers with a smarter, more secure and efficient ecosystem of biometric enrolment options, for an improved customer experience.
The Home Office’s ambition is that all visitors and migrants will provide their biometric facial images and fingerprints under a single global immigration system ahead of travel to the UK, utilising remote self-enrolment for those who are not required to apply for a visa.
BSEFT has been developed by Joint Security and Resilience Centre (JSaRC) on behalf of Future Borders and Immigration System (FBIS) and in partnership with colleagues from Home Office Biometrics (HOB), Identity Security Policy and DDaT’s User Centred Design team.
These trials are at the forefront of the Home Office’s transformational journey which have been developed to assess and drive the market for smartphone fingerprint capture.
The first round of trials was held in Manchester in late 2021; involving 14 suppliers and 545 participating members of the public:
As the results from this revealed that this technology required further development and maturity, the project focused on continuing to drive this capability though internal benchmarking trials. The first of these took place in November 2022, testing the capability of smartphone App solutions from 5 leading industry suppliers. Results were assessed and analysed, and individual supplier performance feedback reports compiled and shared to enable iterative technological development.
The second round of benchmarking trials were held in Sept 2023, and involved 9 technology suppliers and over 120 Home Office volunteers, enabling them to try out a range of different industry solutions, providing a unique view of the application process through the lens of a customer. Each app was assessed against a range of performance criteria, including on both Android and IOS devices. There were also tests for self-enrolment vs assisted enrolment, this is where the participant’s fingerprints are captured by another person in control of the capture device (smartphone)
There were three main pre-defined success criteria themes being evaluated during trials, these included.
- HOB Technical Criteria
- Verification: Confirming an identity claim through biometric comparisons. Am I who I say I am? (1:1) (See further details: biometricsinstitute.org)[1]
- Identification: Search against a biometric enrolment database to find and return the biometric reference attributable to a single individual. Is the data subject in the database? (1:N) (See further details: biometricsinstitute.org)
- Completion rates the ability to successfully complete a transaction to capture all available fingerprints.
- Usability Criteria – User satisfaction levels.
- Presentation Attack Detection (PAD) – Ability to detect genuine/non-genuine attempts to subvert the enrolment system.
2023 Trials highlights included:
- 4 days of trials.
- iOS and Android App versions tested.
- Over 10,000 fingerprints captured.
- 11 hours of user research interviews.
- Over 124 hours of testing by trial participants.
Results:
Analysis of September 2023 Benchmarking trials data confirmed strong areas of development across a number of biometric self-enrolment solutions, providing positive indication of maturing market:
- Technical Performance (HOB) is significantly closer to target:
- Results show some suppliers are achieving or very close to target for verification, identification, and completion rates.
- 4 suppliers achieved 95% ten fingerprint completion rate, within three attempts.
- All suppliers achieved the FAR (false accept rate) target, One achieved FRR (false reject rate) target.
- Of 4 suppliers that participated in both BM1 & 2, all 4 were able to demonstrate an improvement in one or more of the areas of success criteria (verification, identification and completion rates).
- Useability has improved significantly:
- More suppliers are performing better across a range of usability features.
- Suppliers involved in BM1 & 2 have shown strong development, confirming the benefits of participation in BSEFT trials.
- Some solutions only require minor developments to achieve broader satisfaction levels.
- PAD (Presentation Attack Detection) continues to improve:
- There were only two suppliers who took part in BM1 & 2 but both showed significant improvement.
- Generally, positive maturity reflected by more suppliers successfully identifying lower sophistication attacks.
- More focus needs to be made of PAD i.e. The number of suppliers offering PAD needs improvement (5/9)
Outcome and Next Steps:
Suppliers App performance for contactless self-enrolment via smartphone have confirmed market maturity with results displaying an improvement in App capability to levels which sufficiently satisfy the pre-determined success criteria, a critical step to delivering the Government’s current ambition for biometric collection.
As these results are based on a relatively small sample of HO users from benchmarking trials, further testing is required to assess how well market solutions perform on a larger and more diverse cohort of participants, including for the first time in BSEFT trials, children.
BSEFT is therefore developing plans for another round of trials in Q4 2024. This will again include members of the public but will for the first-time target people who are much closer to the intended use case for this technological capability i.e. those who’ve been involved in existing immigrations systems and processes and experienced some of its challenges. It will also aim to include a degree of remote testing, where users download and test suppliers App in their own homes, on their own devices, and involving their family.
This is intended to provide unprecedented insight into the intended user experience and offer assurance that this technology is ready for pilot implementation.