[09-10]aml analysis-id verification online

Local accumulated smoothing patterns for fingerprint liveness detection

Date:2021-09-10

video kyc vendors

Author:cdd anti money laundering    video kyc solution

Keywords:jumio pricing,    pii data protection,    kyc meaning in bank account,    age verification,    define biometric authentication,  kyc passport verification

Description:

On Improving Temporal Consistency for Online Face Liveness Detection
Biometric systems are emerging technologies that enable the authentication of an individual based on physiological or behavioral characteristics. Biometric techniques include recognizing faces, fingerprints, hands geometry, palms, voices, gait, irises, signature, etc.
Face verification can be used in various fields like image and film processing, human-computer interaction, forensic applications etc. Spoofing using photographs or videos is a major problem in face verification systems. A face verification system has basically two parts (1) a face recognition system (2) liveness detection system. Researches to develop simple and easy to implement computational models to identify the faces and antispoofing techniques has been progressing. Although the face images have high dimensionality, they actually span very low dimensional space. In order to find the lower dimensional space, the Eigen face approach, that uses Principal Component Analysis (PCA) algorithm for the recognition of the images, is used in this paper. The goal is to implement the system (model) for a particular face and distinguish it from a large number of stored faces. To address the spoofing problem a real-time and nonintrusive method based on the diffusion speed of a single image is proposed. The antispoofing features are obtained based on the observation that the difference in surface properties between a live face and a fake one is efficiently revealed in the diffusion speed. More specifically, the features used here are the local patterns of the diffusion speed, the so-called local speed patterns, which are input into the linear SVM classifier to classify the given face as fake or live. Advantages of the proposed method in contrast to previous approaches are it performs well regardless of the image medium and even under varying illuminations. No specific user action is required.

Similar Articles