Capture readiness
Module status, palm position and illumination are checked before biometric processing begins.
Veia da palma / PAD testing
Presentation attacks meet their match.
BioWavePass evaluates palmprint and palm vein authentication against common and advanced spoofing methods—combining dual-spectrum imaging with layered security for payment scenarios.
✋
RGBSurface detail
NIRVein signal
● LIVE PALM SIGNAL
Internal Presentation Attack Detection testing focused on real-world payment authentication.
8 attack methods
RGB + NIR dual-spectrum sensing
1M+ conventional spoof records*
Extremely low overall assessed risk*
01PAD methodology
A real palm has more than one story to tell.
The device incorporates dual-spectrum RGB and NIR liveness detection technology to mitigate presentation attacks such as photos, video replays and prosthetic replicas. Each capture must progress through multiple independent checks.
Liveness + image quality
Surface and sub-surface observations help distinguish living tissue from displayed, printed or fabricated inputs.
Alignment + reliability
Palm geometry is aligned and sample reliability is assessed before recognition or registration.
Dual-modal verification
Palmprint texture and palm vein features are evaluated alongside enrollment and account controls.
02Attack coverage
Eight ways to imitate a palm.
Multiple ways to detect the difference.
All attacks Lower risk Medium risk Higher risk
Printed media
Printed palmprint
RGB paper images do not reproduce sub-surface vein information.
Low risk
< $1
< $1
Printed media
Printed palm vein
Printed infrared imagery lacks corresponding surface palmprint detail.
Low risk
< $1
< $1
Composite
Paper splice
Adhesive traces and aliased print/vein signals expose mixed samples.
Low risk
Low cost
Low cost
Replay
Screen replay
Visible screen light cannot reproduce near-infrared vein imaging.
Very low
< $100
< $100
Fabricated
3D palm model
Material response and missing dynamic vascular signals distinguish prosthetics.
Med–high
Up to $20k
Up to $20k
Wearable
Glove attack
Material differences remain observable; transparent gloves reveal the wearer.
Médio
Variable
Variable
Reconstruction
Feature restoration
Image-to-feature extraction is irreversible; encrypted transport adds protection.
Médio
High cost
High cost
Biological material
Animal skin
Dried animal skin differs from living human tissue and lacks vascular dynamics.
Médio
Very high
Very high
03Algorithm flow
Quality first. Then recognition.
The palm registration and recognition workflow shows how capture controls, liveness, alignment and reliability decisions connect before feature extraction.
Palm registration + recognition algorithm Click the diagram for a full-size view
04Attack economics
Imitation gets expensive—fast.
A more realistic fake demands specialist materials, target biometric data and success across every security checkpoint. In most payment scenarios, cost rises faster than likely return.
Paper attack < $1
Screen replay < $100
Ordinary silicone prosthetic $100–500
High-precision 3D prosthetic $5k–20k
Dynamic biomimetic fake > $50k
Estimated attack-cost bands from the BioWavePass internal assessment. Values are indicative and scenario-dependent.
Assessment conclusion
Layered verification makes practical spoofing commercially unattractive.
Common print, replay, splice and ordinary prosthetic attacks are addressed by established defensive coverage. Advanced attacks remain theoretically possible, but must satisfy liveness, palmprint, palm vein, image quality, enrollment and account checks at the same time.
* Findings summarize “Testing of Spoofing Attack Methods — Focusing on Payment Scenarios,” an internal BioWavePass assessment. ISO/IEC 30107-3 is referenced only as a methodology framework informing algorithm training and internal PAD testing. BioWavePass does not claim ISO/IEC 30107-3 certification, formal conformance or third-party PAD evaluation. Results are scenario-dependent and are not an absolute guarantee against every attack.
↓
Overall assessed risk
Extremely low
Overall assessed risk
Extremely low