A Novel Privacy Preserving Biometric Authentication Scheme Using Polynomial Time Key Algorithm In Cloud Computing

Tumuluru, Praveen; Burra, Lakshmi Ramani; BhavaniDasari, Durga; Saibaba, Ch.M.H.; Revathi, B.; Venkateswarlu, Bondu

doi:10.1109/icais50930.2021.9395964

Cited by 7 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This system recommended several privacy-preserving biometric verification solutions. Current research has mostly concentrated on the secure administration of biometric templates (Tumuluru et al, 2021). Despite the fact that the template reduces the likelihood of a direct leak of the original information, once the template is leaked, the original information can be roughly ascertained by reverse engineering (Lee & Jeong, 2021).…”

Section: Introductionmentioning

confidence: 99%

Facial and Palm-Based Biometric Authentication and Data Security

Tiwari,

Jha

2024

Preprint

View full text Add to dashboard Cite

Biometric Authentication (BA) is frequently used for authentication owing to its high recognition rate. The existing biometric hiding algorithms execute data embedding on areas that do not encompass key features of the biometric. Moreover, these techniques lacked authorization. Thus, a Secure Data Transfer model with BA and Blockchain (BC)-based authorization is proposed. Primarily, the data owner registers their details and the registered face and palm image undergoes pre-processing. By employing Pruned Residual Network 50 (PRESNET 50), the facial landmarks are extracted from the pre-processed face image. Next, Digit Folding based Log Facial Jaw Points Curve Cryptographic (DF-LFJPCC) is executed based on the jaw points to generate a secret key. Then, the Tan Sigmoid-based Convolutional Neural Network (TS-CNN) classifier is trained with the features of the pre-processed images and facial landmarks. After registration, the user logins, and their processed face and palm features are given to the trained TS-CNN for authenticating the user. The secret is also used to improve the authentication process. After successful login, the file to be uploaded is converted into cipher, which is then encrypted using Log Facial Jaw Points Curve Cryptographic (LFJPCC) and uploaded to the cloud server. In the end, authorization is performed in the BC based on the hashcode generated using Faro shuffle -Tiger (FS-Tiger) when a user requests data. As per the experimental analysis, the proposed technique outperforms prevailing models.

show abstract

Section: Introductionmentioning

confidence: 99%