There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Nowadays, there is an increasing tendency to fuse multi-modal medical image due to the high diagnostic accuracy and assessment. However, the distribution of these images and more generally, health records may bring serious copy-violation issues in healthcare applications. Further, a massive amount of electronic health records is uploaded on the cloud due to the low cost and high usability. However, outsourcing these records to the cloud may greatly increase security and privacy issues. The present article develops medical image fusion-based watermarking scheme in dual tree complex wavelet transform (DTCWT)-singular value decomposition (SVD) domain. Here, computed tomography (CT) and magnetic resonance imaging (MRI) images are partitioned into a sequence of high frequencies and low frequency by non-subsampled shearlet transform (NSST) to obtain a fused image as a carrier. To enhance the authenticity, spatial and transform domain-based embedding is conducted to conceal multi-marks into carrier image. Finally, improved encryption scheme is employed to obtain a secure marked carrier image. Objective and subjective evaluations indicate a strong robustness effect (43.64%) of the proposed algorithm compared to other methods with remarkable invisibility, payload, and security. To the best of our knowledge, studies of image fusion-based dual watermarking and encryption have not yet been developed in combination with spatial and transform domain schemes for copy protection and ownership of health records, since it is also suitable for applications in the cloud. 

Health Record Security Through Multiple Watermarking on Fused Medical Images

Introduction Information hiding is an idea that’s often cited as the foundation of modern software design methods. Many developers practice it in some form, either intuitively and unable to say exactly what they are doing or how they do it, or knowingly and systematically. This paper is intended for software designers, developers, and managers who agree with the objectives of information hiding, but who may have difficulty giving a precise statement of the information hiding principle, or saying exactly how they applied the information hiding principle in their designs and code, or articulating the benefits and risks of applying the principle. For the architect/designer who is wondering how to create a design that results in software that is easy to change, easy to understand, and easy to verify one piece at a time, this paper provides some thoughts and guidance on how information hiding can help. It also serves as an introduction for those who have heard the term information hiding, but have never had a clear idea of exactly what it means.

Information Hiding

This paper provides a multimodal biometric watermarking algorithm to protect and well authenticate biometric data. It uses the DTCWT-DCT as a combination of two domain transform techniques to fuse face and fingerprint modalities. Two pseudorandom sequences are used to embed the bits of the quantized spectral minutiae representation used as a watermark into the face image. The embedding is done only in high-frequency sub-bands of the DTCWT decomposition, after applying the DCT transform to improve the authentication system security. The conducted experimental results based on ORL face and FVC2002 DB1 fingerprint databases proved the ability of the proposed approach to withstand the conventional digital image watermarking attacks. Also, good performance is showed in term of attaining an accuracy rate of 100%.

DTCWT-DCT watermarking method for multimodal biometric authentication

Due to the rapid advancement of technology, biometrics is widely used in authenticating a person. The primary issue in biometric authentication is template protection, which protects the user’s privacy. To achieve this level of security, we transform the original template generated from the extracted features into a pseudo template. This paper considers two biometric traits, namely Iris and Fingerprint. First, the features are extracted from both the traits. After extracting the features, the feature vectors are compressed to reduce the size. Then the feature vectors of both the traits are fused. Later, local random projection (LRP) is applied to this template to build a revocable and un-linkable template. Finally, fully homomorphic encryption (FHE) is applied to this generated template to secure the user’s privacy, as all operations on the template are carried on encrypted data. Our proposed method improves the accuracy of the system as it generates rotational invariant codes. This proposed method is tested on the Children Multimodal Biometric Database (CMBD), which shows that our method can maintain user privacy with a good accuracy rate. Our method achieves an EER of 0.0214%. 

Securing multimodal biometric template using local random projection and homomorphic encryption

In the last few decades, deep-learning-based face verification and recognition systems have had enormous success in solving complex security problems. However, it has been recently shown that such efficient frameworks are vulnerable to face-spoofing attacks, which has led researchers to build proficient anti-facial-spoofing (or liveness detection) models as an additional security layer. In response, increasingly challenging and tricky attacks have been launched to fool these anti-spoofing mechanisms. In this context, this paper presents the results of an analytical study on transfer-learning-based convolutional neural networks (CNNs) for face liveness detection and differential evolution-based adversarial attacks to evaluate the efficiency of face anti-spoofing classifiers against adversarial attacks. Specifically, experiments were conducted under different use-case scenarios on four face anti-spoofing databases to highlight practical criteria that can be used in the development of countermeasures to address face-spoofing issues.

Unraveling robustness of deep face anti-spoofing models against pixel attacks

https://www.techscience.com/iasc/v27n1/41145/pdf

Hybrid Multimodal Biometric Template Protection

This paper presents a spatial biometric watermarking scheme for multimodal authentication system. This method embeds the face features (watermark) into the fingerprint image (cover) via a generated secret key to locate the pixels to be watermarked while preserving the fingerprint minutia points. The face features are extracted using Orthogonal Locality Preserving Projections technique. The proposed scheme has the advantage that in addition to fingerprint matching, the recovered watermark during the decoding can be used to establish the authenticity of the fingerprint and the user. The experimental results obtained using the ORL face database and FVC2002 DB2 fingerprint database have showed that the fingerprint-face based multimodal authentication system has good performance accuracy.

Watermarking for protected fingerprint authentication

AI-based face recognition has become increasingly appealing for daily life applications due to its high performance. At the same time, image coding is very commonly used and, thus, many applications perform face recognition with decoded images. However, using decoded images, which may suffer from compression artifacts, may impact the final decision-making process of AI-based face recognition systems and, thus, its overall recognition performance. This paper studies the impact of image coding on the overall face verification performance of a popular and high performing face recognition solution, ArcFace. Face recognition using both original and decoded images, with several compression rates and qualities, is considered. Tests were performed using the Labeled Faces in the Wild (LFW) face dataset, with its images coded using conventional image coding standards, notably JPEG, JPEG 2000, and JPEG XL, as well as three emerging AI-based image codecs. As expected, the experimental results show that coding can have a significant impact on face recognition performance, with its impact becoming increasingly relevant as the coding rate is reduced. It is also observed that the recent AI-based image codecs appear to offer slightly better recognition performance for the same coding rates as a consequence of their better RD performance.

Naima Bousnina

Papers

Unraveling robustness of deep face anti-spoofing models against pixel attacks

DTCWT-DCT watermarking method for multimodal biometric authentication

Hybrid Multimodal Biometric Template Protection

Watermarking for protected fingerprint authentication

Impact of Conventional and AI-based Image Coding on AI-based Face Recognition Performance