Palm Vein Technology
PalmPay reads the vascular pattern inside the palm with near-infrared light, isolates it on the device, and converts it into an irreversible template. The raw image is destroyed at the edge. Everything below is the chain that makes that possible.
Where the work happens
Recognition runs on the terminal itself. That is an architectural choice with two consequences a reviewer should hold us to: nothing that could identify a person has a network path off the device, and a site with an unreliable uplink still verifies people in under a second.
Why now
None of this is our own telemetry. Each figure below is publicly reported, with the source stated under it.
~10x
rise in reported cyber-fraud losses in India, 2022 to 2024
Ministry of Home Affairs, data tabled in Parliament
85%
year-on-year increase in domestic UPI fraud cases, FY23 to FY24
Finance Ministry, Lok Sabha reply, November 2024
47%
of Indian adults have experienced, or know someone who has experienced, an AI voice or deepfake scam
McAfee, 2023 consumer survey
Reported cyber-fraud losses
Losses more than tripled between 2022 and 2023, and roughly tripled again between 2023 and 2024. That is a trend line moving faster than manual verification can keep up with.
Ministry of Home Affairs, data tabled in Parliament
The alternatives
Face, iris and fingerprint are not wrong technologies. Each solves a real problem elsewhere. None of them was built for a device-free, high-throughput, industrial-grade Indian deployment.
01
A face is a public, outward-facing credential: it is on every ID card and photo online, and now reproducible by generative tools that cost under $50 a month. Liveness checks built to catch a mask or a printed photo are being defeated by models that replicate blink and micro-movement, the exact deepfake pattern Indian banks are now seeing at onboarding.
02
Accurate, but it needs a cooperative subject, a controlled distance and near-infrared illumination precise enough that dust, glare or a tilted head causes a rejection. Expensive to deploy at a shop-floor gate or a retail counter, and not built for the throughput a shift change or a queue demands.
03
The credential lives on the surface: it wears down with manual labour, degrades with grime, moisture and callusing, and, unlike a vein pattern, can be lifted and replicated from a print left on any surface the finger has touched.
04
The pattern sits beneath the skin and needs live blood flow to read at all, so it is not something a photograph, a mould or a lifted print can reproduce. Unaffected by surface wear, grime or moisture, and the same contactless gesture works at a checkout counter and a factory turnstile.
We will walk your security, compliance and operations teams through the pipeline, the template format and the pilot deployment plan, and answer the questions these pages do not.