English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Fingerprint authentication is a popular security mechanism for smartphonesand laptops. However, its adoption in web and cloud environments has beenlimited due to privacy concerns over storing and processing biometric data onservers. This paper introduces Blind-Touch, a novel machine learning-basedfingerprint authentication system leveraging homomorphic encryption to addressthese privacy concerns. Homomorphic encryption allows computations on encrypteddata without decrypting. Thus, Blind-Touch can keep fingerprint data encryptedon the server while performing machine learning operations. Blind-Touchcombines three strategies to efficiently utilize homomorphic encryption inmachine learning: (1) It optimizes the feature vector for a distributedarchitecture, processing the first fully connected layer (FC-16) in plaintexton the client side and the subsequent layer (FC-1) post-encryption on theserver, thereby minimizing encrypted computations; (2) It employs a homomorphicencryptioncompatible data compression technique capable of handling 8,192authentication results concurrently; and (3) It utilizes a clustered serverarchitecture to simultaneously process authentication results, therebyenhancing scalability with increasing user numbers. Blind-Touch achieves highaccuracy on two benchmark fingerprint datasets, with a 93.6% F1- score for thePolyU dataset and a 98.2% F1-score for the SOKOTO dataset. Moreover,Blind-Touch can match a fingerprint among 5,000 in about 0.65 seconds. With itsprivacyfocused design, high accuracy, and efficiency, Blind-Touch is apromising alternative to conventional fingerprint authentication for web andcloud applications.

Blind-Touch: Homomorphic Encryption-Based Distributed Neural Network Inference for Privacy-Preserving Fingerprint Authentication