Palm Verification
Palm Verification is a way to authenticate online a living person physically while maintaining their privacy. No special hardware is needed to use the functionality, just a common smartphone. The feature provides client components for auto capture of the palm photo. The functionality on the server provides palm image comparison and verification.
A person needs to take image of their palm when creating an account that uses palms for verification. This is stored on the server to be used for comparison against future images taken for verification.
Benefits
The benefits of the palm biometrics modality compared to face are:
- Security - images of a person’s palms are not available on social media.
- Privacy - images of palms cannot be matched against facial images or ID documents. Person uses his biometrics without compromising privacy.
- Consent - taking image of one’s palm requires consent. The person needs to present its palm to camera, it can be hardly taken from a person walking on the street.
- Accuracy - a palm carries more information than a fingerprint or iris.
Palm Photo Autocapture
Palm auto capture is a client component to capture an image of a palm suitable for verification without having to manually trigger the photo capture. Image is captured automatically when all quality requirements have been satisfied.
Palm auto capture is provided by the following components of mobile app libraries:
For web development it’s the Palm auto capture web component:
- Web Palm Auto Capture component (to be released in Q4 2024)
Palm Comparison and Storage
In order to compare two palms, the following steps are performed:
- Template extraction - detect the palm in the image and compute its representation for comparison
- Comparison - compare two palm templates and return a similarity score
Template extraction
Palm template is a set of stored biometric features extracted from the image. This can be used for comparison and is usually in a form of binary data that has a significantly smaller size than the original image. These templates are stored or cached in the solution to perform the comparison. Once a reference template is stored on the server, the extraction is performed only on the probe image from the new authentication. Only templates generated by the same mode and product version can be matched. During some product upgrades, templates must be regenerated as mentioned in the respective product changelog.
Comparison (formerly Matching)
Comparison calculates the similarity of the two templates, providing a comparison score. The higher the score, the more similar are the palms. This operation is usually very fast.
The Digital Identity Service (DIS) provides the palm comparison functions as dedicated palm API functions.
- DOT Digital Identity Service - Palms API
- API documentation: palms/{id}/similarity
Comparison threshold and comparison decision
The final comparison decision if the two palm images belong to the same person and same hand is determined by the comparison score and the threshold. If the score is above the threshold this can be interpreted as a match, if the score is below the threshold it is a non-match.
Comparison accuracy
Comparison thresholds affects the decision mistakes. False match is when comparison decides that palm images of two different hands are matching. False Match Rate FMR (formerly FAR) is the proportion of comparison trials that result in a false match on a given dataset with a given threshold. False non-match is when comparison decides that two palm images of the same hand do not match. False Non-match Rate FNMR (formerly FRR) is the proportion of comparison trials that result in a false non-match on the same dataset and with the same threshold. The combination of these two FMR @ x% FNMR defines the accuracy of comparison.
Threshold for comparison
We recommend using the threshold 0.5 for the comparison score to declare match between two palms.