01EnrollmentIdentity + biometric
BIOMETRIC MATCHING
Biometric Verification 1:1, Identification 1:N & Deduplication
Separates objectives, galleries, metrics, decisions and evidence so verification, authentication and identification are not conflated.
ENGINEERING COMPARISON
Different from input through evidence
Treats each matching task as a distinct technical operation.
| Dimension | 1:1 Verification | 1:N Identification | Deduplication |
|---|---|---|---|
| Question | Is this the claimed identity? | Who is this probe in the gallery? | Has this identity already been enrolled? |
| Input | Probe + known reference | Probe + gallery | Enrollment probe + enrolled population |
| Output | Score / accept-reject | Candidate list / rank / threshold | Duplicate candidates / review |
| Metrics | FMR · FNMR · FAR/FRR concepts | FPIR · FNIR · Rank-N · gallery size | False candidate management · review workload |
| Evidence | Reference ID · score · threshold · policy | Gallery/version · candidates · rank · threshold | Enrollment context · candidate review · decision |
IDENTIFICATION ARCHITECTURE
ABIS-style architecture without unsupported product claims
Unless product evidence explicitly classifies CheckID as an ABIS, the public name remains Biometric Identification Architecture.
02QualitySensor / capture quality
03TemplateProtected biometric reference
04GalleryVersioned reference set
05Search1:N engine
06Candidate ListRank / score / threshold
07AdjudicationPolicy / reviewer if applicable
08EvidenceDecision context
BIOMETRIC IDENTIFICATION
Verification, identification and deduplication solve different problems
Uses professional 1:1 / 1:N terminology with gallery, candidate, rank, threshold and adjudication concepts.
verify
1:1 Verification
Compare a probe against the biometric reference of a claimed identity.
InputProbe + claimed identity
OutputScore + decision
verify
1:N Identification
Search a gallery and return ranked candidates according to search policy.
InputProbe + gallery
OutputCandidate list + rank
exception
Deduplication
Search before enrollment to identify a possible prior identity.
InputNew enrollment
OutputCandidate / no candidate
assure
Adjudication
Policy or human review may resolve ambiguous candidates; reference architecture unless implemented.
InputCandidates + evidence
OutputConfirm / reject / escalate
ABIS-STYLE REFERENCE
Logical identification architecture
The architecture is presented professionally without claiming CheckID is a full ABIS product unless evidence confirms it.
01
Capture
Face · fingerprint · iris · palm where supported.
02
Quality Gate
Reject or reacquire poor-quality captures before extraction/search.
03
Template
Generate engine-specific biometric reference.
04
Gallery
Index/partition enrolled biometric references.
05
Matcher
Perform 1:N search and generate candidate scores/ranks.
06
Adjudication
Policy/human review where required.
07
Evidence
Record context, engine/version, result, policy and audit.
Technical basisSOURCE-BACKED
SourceCHECKID source register
StandardRoute-specific standard mapping
Version
ApplicabilitySee page scope
EvidenceQA evidence register
TERMINOLOGY
Biometric identification is not e-Identification
1:1 verification · 1:N biometric identification · electronic identification
1:1 biometric verification
Compare the current sample with the reference of a known subject.
1:N biometric identification
Search a gallery for candidates; gallery scope, thresholds and adjudication semantics are required where supported.
e-Identification
Electronic identification at the identity-data/eID layer, not a 1:N search.