Systems and methods for performing fingerprint based user authentication using imagery captured using mobile devices

Inventors

Othman, Asem • Tyson, Richard • Tavanai, Aryana • Xue, Yiqun • Simpson, Andrew

Assignees

Veridium IP Ltd

Abstract

Technologies are presented herein in support of a system and method for performing fingerprint recognition. Embodiments of the present invention concern a system and method for capturing a user's biometric features and generating an identifier characterizing the user's biometric features using a mobile device such as a smartphone. The biometric identifier is generated using imagery captured of a plurality of fingers of a user for the purposes of authenticating/identifying the user according to the captured biometrics and determining the user's liveness. The present disclosure also describes additional techniques for preventing erroneous authentication caused by spoofing. In some examples, the anti-spoofing techniques may include capturing one or more images of a user's fingers and analyzing the captured images for indications of liveness.

Core Innovation

The invention relates to fingerprint recognition using a mobile device having a camera. Images depicting a plurality of fingers of a subject are captured, finger(s) are detected from the images using a finger detection algorithm, and fingertip segments are identified from the images using a segmentation algorithm. A convolutional neural network (CNN) assesses a quality score of the at least one image, and the identified fingertip segment is used to measure fingerprint-ridge-related features.

The invention includes measuring one or more features of the at least one finger, wherein the measured feature is a frequency of fingerprint ridges. The at least one image is scaled based on the measured frequency of fingerprint ridges and one or more of a prescribed reference frequency and a target resolution, and a biometric identifier is generated that comprises at least a portion of the scaled image depicting the respective fingertip segment.

The invention also includes extracting features and minutia points from the identified fingertip segment(s), where the quality score guides subsequent processing. A CNN trained to detect minutia points detects minutia points within at least one image, determines respective quality scores for the detected minutia points, and selects a subset of the minutia points for inclusion in the biometric identifier according to the respective quality scores. In addition, minutia points are detected using a minutia extraction algorithm, quality scores are calculated for minutia points, and at least a subset of the minutia points is selected for inclusion in the biometric identifier.

Claims Coverage

The partial content contains three independent claims, each describing a mobile-device fingerprint recognition method that captures finger images, detects and segments fingers into fingertip segments, assesses image quality using a CNN, and generates and stores a biometric identifier. The inventive features across the independent claims differ primarily in ridge-frequency-based scaling tied to a target resolution and reference frequency, and minutia-point detection and quality-based selection pipelines, including selection guided by minutia quality scores.

Across the independent claims, the core inventive coverage centers on generating biometric identifiers from mobile-captured finger images by detecting fingers, segmenting fingertip regions, assessing image quality using a CNN, and extracting fingerprint features. One independent claim additionally emphasizes scaling fingertip images based on measured fingerprint-ridge frequency using a prescribed reference frequency and target resolution, while the other two independent claims emphasize detecting minutia points and selecting a subset based on minutia quality scores for inclusion in the biometric identifier.

Stated Advantages

Documented Applications