Tool for assessing biometric samples in various modalities launched by Biometix
Biometix has launched a new tool for assessing the quality of biometric data from the four most common physical modalities in various ways.
The new Biometric Quality Assessment Tool (BQAT) provides analysis of fingerprint, face, iris, and voice biometrics samples.
Fingerprint analysis is based on NIST and NFIQ2 quality features, and links image quality to recognition performance, whether with optical or ink 500 PPI samples. Face image assessment analyzes criteria like head pose, smile detection, inter-eye distance, and open or closed eyes. Iris analysis refers to ISO metrics and various quality attributes, and speech samples are examined for characteristics like naturalness, coloration and noisiness.
“BQAT (Biometric Quality Assessment Tool) is a framework to generate and analyze quality metrics of biometric samples to meet international standards as well as support for customized metrics,” says Biometix CEO Ted Dunstone wrote in a LinkedIn post. “BQAT can, for instance, take an input directory of biometric data and produce both the raw quality information as well as an analysis report and outlier detection.”
Biometix says installation is simple through a docker image, or a local build with a git clone, and BQAT is designed for easy integration with existing systems. The tool’s source code is also publicly available to allow for forking.
Dunstone says further announcements about the application of BQAT are coming soon.