FIGURE 5 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Main steps involved in the feature extraction: (a) ROI image, (b) blood vessel enhancement through contrast and sharpness adjustment, (c) highlighting line-like features and fine details using the LoG, and (d) final result after morphological operations.
Source publication
Securing personal information and data has become an imperative challenge, especially after the introduction of legal frameworks, such as, in Europe, the General Data Protection Regulation (GDPR). Conventional authentication methods, such as PINs and passwords, have demonstrated their vulnerabilities to various cyber threats, making it necessary th...
Context in source publication
Context 1
... suggested feature extraction process encompasses the following steps: 1) emphasising blood vessels through contrast and sharpness enhancement, 2) employing Laplacian of Gaussian, 3) using morphological operations. Figure 5 illustrates the results of the above-mentioned steps whose details are given below. ...
Similar publications
The face is one of the body parts that a detection system can detect. This is what underlies the existence of face detection. One technology that uses face identification is biometric identification. The Viola-Jones method determines whether an object is a face by extracting features in a face image and classifying it to decide whether or not it is...
Citations
... Retinal data remained encrypted throughout the process, minimizing risks of breaches. While the system ensured robust privacy and security, faced challenges related to encrypted data management and computational overhead [64]. ...
The widespread adoption of smart technologies and digital services has intensified the demand for reliable and secure identity authentication systems. Biometric credentials, offering significant advantages over traditional methods, have emerged as a preferred solution. However, preserving the privacy of biometric data remains a critical challenge. This paper presents a comprehensive study of privacy-preserving techniques and potential risks associated with biometric systems. A novel taxonomy and structured framework are introduced to classify existing research in this domain. The study evaluates state-of-the-art approaches across a wide range of biometric modalities, including facial recognition, fingerprint analysis, eye-based identification, Electroencephalogram (EEG), Electrocardiogram (ECG), voice recognition, and hand geometry, highlighting their respective benefits, limitations, and application areas. Additionally, the paper examines the balance between privacy, security, and recognition performance, addressing current challenges and proposing future research directions. This work serves as a valuable resource for researchers and industry professionals, supporting the development of privacy-centric and secure biometric authentication technologies in an evolving digital landscape.
Carpooling enables multiple users with similar travel habits to share rides, reducing vehicles on the road, leading to benefits such as lower fuel consumption, reduced traffic congestion, and lower environmental impact. However, carpooling also poses a challenge to the privacy of the users, as they may not want to reveal their location or route information to others. This research study delves into a cutting‐edge approach to address these privacy concerns by leveraging homomorphic encryption (HE) within the realm of carpooling services. The proposed solution makes use of a HE scheme that supports encrypted computation on real numbers, which is suitable for carpooling applications that involve distance and time calculations. The approach enables decision makers to perform efficient and accurate route matching over encrypted data, without disclosing their sensitive information about users, thus preserving the confidentiality of the data. The proposed system is evaluated through extensive experiments and simulations, demonstrating its effectiveness in terms of both security and privacy when the system operates in normal (ideal) and abnormal (under attack) conditions. Experimental results indicate that the proposed solution offers robust resistance to various attacks, including replay attacks and data exposure, providing a robust and privacy‐centric solution for carpooling services.
