Researching a machine learning algorithm for a face recognition system

This article investigated the problem of using machine learning algorithms to recognize and identify a user in a video sequence. The scientific novelty lies in the proposed improved Viola-Jones method, which will allow more efficient and faster recognition of a person's face. The practical value of the results obtained in the work is determined by the possibility of using the proposed method to create systems for human face recognition. A review of existing methods of face recognition, their main characteristics, architecture and features was carried out. Based on the study of methods and algorithms for finding faces in images, the Viola-Jones method, wavelet transform and the method of principal components were chosen. These methods are among the best in terms of the ratio of recognition efficiency and work speed. Possible modifications of the Viola-Jones method are presented. The main contribution presented in this article is an experimental study of the impact of various types of noise and the improvement of company security through the development of a computer system for recognizing and identifying users in a video sequence. During the study, the following tasks were solved: – a model of face recognition is proposed, that is, the system automatically detects a person's face in the image (scanned photos or video materials); – an algorithm for analyzing a face is proposed, that is, a representation of a person's face in the form of 68 modal points; – an algorithm for creating a digital fingerprint of a face, which converts the results of facial analysis into a digital code; – development of a match search module, that is, the module compares the faceprint with the database until a match is found

Transformation of Mechanisms of Intellectual Rights Protection in the Context of Development of Virtual and Augmented Reality Technologies

The paper examines the main mechanisms for protecting intellectual rights and liability for their infringement when using virtual and augmented reality technologies. The author notes that the market for these technologies is still experimental. Analyzing the prognostic information of experts in the field under consideration, the author identifies the most effective technical means of protection, as well as mechanisms of responsibility for infringement of intellectual rights in the implementation of these technologies. Digital fingerprint technologies, digital marking of works, as well as copyright traps are included into technical means of protecting intellectual rights. The author highlights difficulties in the implementation of jurisdictional forms of protection of intellectual rights when they are infringed in virtual and augmented reality. In this regard, it is concluded that it is necessary to form an electronic justice system to consider disputes concerning intellectual rights in the field of virtual and augmented reality.

Comparison of Two Generations of Systems for Digital Occlusion Examination

BACKGROUND: The modern concept of occlusion includes the relationship between teeth, masticatory muscles, and temporomandibular joints in function and dysfunction. Occlusion can be defined very simply: it means the contacts between teeth. Qualitative and quantitative methods are used to register and evaluate occlusal contacts. The T-Scan handpiece model was updated in 2015 as T-Scan Novus (software version 9.1) and the latest updated one being the T-Scan version 10 software introduced in 2018. AIM: The purpose of this study is to demonstrate the capabilities and results of two generations of systems - T-Scan III and T-Scan Novus. MATERIALS AND METHODS: For the realization of the set goal, the occlusion of a patient with the initials S.K. is examined with two systems. The patient is 43 years old with intact teeth, Angle’s class I jaw relation. The study with T-Scan III was conducted in 2015 and with T-Scan Novus in 2019. RESULTS: The software of both systems uses a graphical interface, which transforms the data obtained during the recording of the occlusion as the model of the upper dentition of the patient in T-Scan III and the upper and lower dentition in T-Scan Novus. Registered occlusal contacts are illustrated as 2D and 3D images of different colors. The graph force/time shows the power versus time from the first contact to the end of the movie. The timing table displays the patient’s total occlusal bite timing, and the force applied. T-Scan Novus software allows you to import digital fingerprint files of the upper and lower dentition in.stl format. CONCLUSION: The software program of the system version 9.1 provides better visualization of dental arches making it much more informative than other versions. The T-Scan system allows fast and accurate registration and analysis of occlusion.

Physically Unclonable Functions with Confidential Computing for Enhanced Encryption of EHRs

Continual exploitation of Electronic Health Records (EHRs) has led to increasing amounts of ransomware and identity theft in recent years. Existing cryptosystems protecting these EHRs are weak due to their inherently transparent software that allows adversaries to extract encryption keys with relative ease. I designed a novel cryptosystem that employs Physically Unclonable Functions (PUFs) to securely encrypt user EHRs in a protected SGX enclave. The CPU-attached PUF provides a secret, device-unique value or a ‘digital fingerprint’ which is used to derive a symmetric key for subsequent AES-NI hardware encryption. Since the cryptographic operations, from key derivation to encryption, transpire in a confidential SGX enclave, the keys are always protected from OS-privileged attacks- a capability lacking in most existing systems. I used my system APIs to evaluate the performance of various hash and encryption schemes across multiple EHR block sizes. SHA512 and AES-NI-256-GCM were selected for cryptosystem implementation because they demonstrated high performance without compromising on security.

Physically Unclonable Functions with Confidential Computing for Enhanced Encryption of EHRs

Continual exploitation of Electronic Health Records (EHRs) has led to increasing amounts of ransomware and identity theft in recent years. Existing cryptosystems protecting these EHRs are weak due to their inherently transparent software that allows adversaries to extract encryption keys with relative ease. I designed a novel cryptosystem that employs Physically Unclonable Functions (PUFs) to securely encrypt user EHRs in a protected SGX enclave. The CPU-attached PUF provides a secret, device-unique value or a ‘digital fingerprint’ which is used to derive a symmetric key for subsequent AES-NI hardware encryption. Since the cryptographic operations, from key derivation to encryption, transpire in a confidential SGX enclave, the keys are always protected from OS-privileged attacks- a capability lacking in most existing systems. I used my system APIs to evaluate the performance of various hash and encryption schemes across multiple EHR block sizes. SHA512 and AES-NI-256-GCM were selected for cryptosystem implementation because they demonstrated high performance without compromising on security.

The Study of Three-Dimensional Fingerprint Recognition in Cultural Heritage

Fingerprints play a central role in any field where person identification is required. In forensics and biometrics, three-dimensional fingerprint-based imaging technologies, and corresponding recognition methods, have been vastly investigated. In cultural heritage, preliminary studies provide evidence that the three-dimensional impressions left on objects from the past (ancient fingerprints) are of paramount relevance to understand the socio-cultural systems of former societies, to possibly identify a single producer of multiple potteries, and to authenticate the artist of a sculpture. These findings suggest that the study of ancient fingerprints can be further investigated and open new avenues of research. However, the potential for capturing and analyzing ancient fingerprints is still largely unexplored in the context of cultural heritage research. In fact, most of the existing studies have focused on plane fingerprint representations and commercial software for image processing. Our aim is to outline the opportunities and challenges of digital fingerprint recognition in answering a range of questions in cultural heritage research. Therefore, we summarize the fingerprint-based imaging technologies, reconstruction methods, and analyses used in biometrics that could be beneficial to the study of ancient fingerprints in cultural heritage. In addition, we analyze the works conducted on ancient fingerprints from potteries and ceramic/fired clay sculptures. We conclude with a discussion on the open challenges and future works that could initiate novel strategies for ancient fingerprint acquisition, digitization, and processing within the cultural heritage community.

Effects of oils, lotions, hand sanitizers, and mehendi on fingerprints captured through digital fingerprint scanner

Background In this modern era, advancement in technology is seen in every aspect of our life making it comparatively much easier. Likewise, in the field of fingerprinting, the digital scanners have replaced conventional methods of taking fingerprints, as it is accurate and less time-consuming. In daily life, people often apply oils, lotions, hand sanitizers, and occasionally mehendi on their hands. These cosmetic and daily use products affect the digital recording of fingerprints, thus making it difficult for forensic experts to identify the real offender in many cases. The purpose of the study was to check the effect of oils, lotions, hand sanitizers, and mehendi on the fingerprint pattern. Results The present study was undertaken by taking 2700 fingerprints from 30 individuals. These fingerprints were recorded with the help of the SecuGen Hamster IV fingerprint scanner under controlled environmental conditions. The examination and comparison of fingerprint patterns were done on the basis of visibility (clarity and intensity). The presence of cosmetic and daily use products affected the visibility of digitally captured fingerprints. Different products caused different effects based on their properties. Synthetic mehendi, alcohol-based hand sanitizer, greasy lotion, and viscous oil caused significant differences in the fingerprint images by degrading the fingerprint quality. The non-greasy lotion and non-alcohol-based hand sanitizer showed less effect, whereas non-viscous oil and natural mehendi caused a minimal effect on the quality of fingerprint images. Conclusion The application of cosmetic and daily use products added an additional layer on the fingers which is not present naturally. The additional layer caused alterations in the fingerprint pattern of an individual. So, digital fingerprints should be collected after proper washing of hands.

A Power Customer Data Relational Algorithm Based on Magnanimity Fuzzy Address Matching

According to the short text and unstructured characteristics of customer address, a data association fusion method for address has been proposed. In this method, the address was mapped to a digital fingerprint by improved Simhash technology, which effectively reduced the dimension of massive addresses and simplified the similarity-matching process of multi-source heterogeneous addresses. Furthermore, the weight setting of the eigenvector of the simhash algorithm was improved by introducing special weight gain. A two-level index mechanism was established by the characteristics of address division and data structure of digital fingerprints; the time-consuming digital fingerprint comparison was greatly reduced. The experimental results showed that calculation efficiency was greatly optimized; accuracy and coverage of the comparison were ensured. Through address matching of different databases, information fusion can be completed and the goal which power customers' demands is connected to power grid equipment is achieved.

Clustering of patient comorbidities within electronic medical records enables high-precision COVID-19 mortality prediction

We present an explainable AI framework to predict mortality after a positive COVID-19 diagnosis based solely on data routinely collected in electronic healthcare records (EHRs) obtained prior to diagnosis. We grounded our analysis on the ½ Million people UK Biobank and linked NHS COVID-19 records. We developed a method to capture the complexities and large variety of clinical codes present in EHRs and we show that these have a larger impact on risk than all other patient data but age. We use a form of clustering for natural language processing of the clinical codes, specifically, topic modelling by Latent Dirichlet Allocation (LDA), to generate a succinct digital fingerprint of a patient’s full secondary care clinical history, i.e. their comorbidities and past interventions. These digital comorbidity fingerprints offer immediately interpretable clinical descriptions that are meaningful, e.g. grouping cardiovascular disorders with common risk factors but also novel groupings that are not obvious. The comorbidity fingerprints differ in both their breadth and depth from existing observational disease associations in the COVID-19 literature. Taking this data-driven approach allows us to avoid human-induction bias and confirmation bias during selection of what are important potential predictors of COVID-19 mortality. Together with age these digital fingerprints are the single most important factor in our predictor. This holds the potential for improving individual risk profiling for clinical decisions and the identification of groups for public health interventions such as vaccine programmes. Combining our digital precondition fingerprints with demographic characteristics allow us to match or exceed the performance of existing state-of-the-art COVID-19 mortality predictors (EHCF) which have been developed through expert consensus. Our precondition fingerprinting and entire mortality prediction analytics pipeline are designed so as to be rapidly redeployable, e.g. for COVID-19 variants or other pre-existing diseases.

Clustering of patient comorbidities within electronic medical records enables high-precision COVID-19 mortality prediction

We present an explainable AI framework to predict mortality after a positive COVID-19 diagnosis based solely on data routinely collected in electronic healthcare records (EHRs) obtained prior to diagnosis. We grounded our analysis on the 1/2 Million people UK Biobank and linked NHS COVID-19 records. We developed a method to capture the complexities and large variety of clinical codes present in EHRs, and we show that these have a larger impact on risk than all other patient data but age. We use a form of clustering for natural language processing of the clinical codes, specifically, topic modelling by Latent Dirichlet Allocation (LDA), to generate a succinct digital fingerprint of a patient's full secondary care clinical history, i.e. their co-morbidities and past interventions. These digital comorbidity fingerprints offer immediately interpretable clinical descriptions that are meaningful, e.g. grouping cardiovascular disorders with common risk factors but also novel groupings that are not obvious. The comorbidity fingerprints differ in both their breadth and depth from existing observational disease associations in the COVID-19 literature. Taking this data-driven approach allows us to avoid human-induction bias and confirmation bias during the selection of what are important potential predictors of COVID-19 mortality. Together with age, these digital fingerprints are the single most important factor in our predictor. This holds the potential for improving individual risk profiling for clinical decisions and the identification of groups for public health interventions such as vaccine programmes. Combining our digital precondition fingerprints with demographic characteristics allow us to match or exceed the performance of existing state-of-the-art COVID-19 mortality predictors (EHCF) which have been developed through expert consensus. Our precondition fingerprinting and entire mortality prediction analytics pipeline is designed so as to be rapidly redeployable, e.g. for COVID-19 variants or other pre-existing diseases.