Efficiency of hybrid algorithm for COVID-19 online screening test based on its symptoms

The coronavirus COVID-19 is affecting 196 countries and territories around the world. The number of deaths keep on increasing each day because of COVID-19. According to World Health Organization (WHO), infected COVID-19 is slightly increasing day by day and now reach to 570,000. WHO is prefer to conduct a screening COVID-19 test via online system. A suitable approach especially in string matching based on symptoms is required to produce fast and accurate result during retrieving process. Currently, four latest approaches in string matching have been implemented in string matching; characters-based algorithm, hashing algorithm, suffix automation algorithm and hybrid algorithm. Meanwhile, extensible markup language (XML), JavaScript object notation (JSON), asynchronous JavaScript XML (AJAX) and JQuery tehnology has been used widelfy for data transmission, data storage and data retrieval. This paper proposes a combination of algorithm among hybrid, JSON and JQuery in order to produce a fast and accurate results during COVID-19 screening process. A few experiments have been by comparison performance in term of execution time and memory usage using five different collections of datasets. Based on the experiments, the results show hybrid produce better performance compared to JSON and JQuery. Online screening COVID-19 is hopefully can reduce the number of effected and deaths because of COVID.

AAU-Net: Attention-Based Asymmetric U-Net for Subject-Sensitive Hashing of Remote Sensing Images

The prerequisite for the use of remote sensing images is that their security must be guaranteed. As a special subset of perceptual hashing, subject-sensitive hashing overcomes the shortcomings of the existing perceptual hashing that cannot distinguish between “subject-related tampering” and “subject-unrelated tampering” of remote sensing images. However, the existing subject-sensitive hashing still has a large deficiency in robustness. In this paper, we propose a novel attention-based asymmetric U-Net (AAU-Net) for the subject-sensitive hashing of remote sensing (RS) images. Our AAU-Net demonstrates obvious asymmetric structure characteristics, which is important to improve the robustness of features by combining the attention mechanism and the characteristics of subject-sensitive hashing. On the basis of AAU-Net, a subject-sensitive hashing algorithm is developed to integrate the features of various bands of RS images. Our experimental results show that our AAU-Net-based subject-sensitive hashing algorithm is more robust than the existing deep learning models such as Attention U-Net and MUM-Net, and its tampering sensitivity remains at the same level as that of Attention U-Net and MUM-Net.

Convolutional Neural Network Copy Detection with Neural Network Perceptual Hashing

In recent years, many model intellectual property (IP) proof methods for IP protection have been proposed, such as model watermarking and model fingerprinting. However, as an important part of the model IP protection system, the model copy detection task has not received enough attention. With the increasing number of neural network models transmitted and deployed on the Internet, the search for similar models is in great demand, which concurrently triggers the security problem of copy detection of models for IP protection. Due to the high computational complexity, both model watermarking and model fingerprinting lack the capability to efficiently find suspected infringing models among tens of millions of models. In this paper, inspired by the hash-based image retrieval methods, we propose a perceptual hashing algorithm for convolutional neural networks (CNNs). The proposed perceptual hashing algorithm can convert the weights of CNNs to fixed-length binary hash codes so that the lightly modified version has the similar hash code as the original model. By comparing the similarity of a pair of hash codes between a query model and a test model in the model library, the similar versions of a query model can be retrieved efficiently. To the best of our knowledge, this is the first perceptual hashing algorithm for CNNs. The experiment performed on a model library containing 3,565 models indicates that our proposed perceptual hashing scheme has a superior copy detection performance.

Convolutional Neural Network Copy Detection with Neural Network Perceptual Hashing

In recent years, many model intellectual property (IP) proof methods for IP protection have been proposed, such as model watermarking and model fingerprinting. However, as an important part of the model IP protection system, the model copy detection task has not received enough attention. With the increasing number of neural network models transmitted and deployed on the Internet, the search for similar models is in great demand, which concurrently triggers the security problem of copy detection of models for IP protection. Due to the high computational complexity, both model watermarking and model fingerprinting lack the capability to efficiently find suspected infringing models among tens of millions of models. In this paper, inspired by the hash-based image retrieval methods, we propose a perceptual hashing algorithm for convolutional neural networks (CNNs). The proposed perceptual hashing algorithm can convert the weights of CNNs to fixed-length binary hash codes so that the lightly modified version has the similar hash code as the original model. By comparing the similarity of a pair of hash codes between a query model and a test model in the model library, the similar versions of a query model can be retrieved efficiently. To the best of our knowledge, this is the first perceptual hashing algorithm for CNNs. The experiment performed on a model library containing 3,565 models indicates that our proposed perceptual hashing scheme has a superior copy detection performance.

Methods and Software Complex Development for Remote Electronic Voting Based on Ethereum Blockchain Platform

The most promising way to increase voters’ confidence in the remote electronic voting (REV) procedure is a voting method based on Ethereum blockchain platform. However, the existing solutions using this method faced a range of problems: ensuring the secrecy of the vote and openness of the procedure for society, pressure on the voter and a guarantee of the reliability of the whole system. In this work, a method for constructing a REV is proposed that solves these problems. It is similar in structure to the traditional voting method, using the same principle and processes. The Ethereum blockchain based REV process is described in detail. It was shown that received votes are securely stored in the Ethereum blockchain network, and the correctness of the vote addressing to the selected candidate can always be checked in real time. The description of smart contract algorithm that implements the transfer of vote from voter to candidate using transactions and determines the winner who received the highest number of votes was provided. It was demonstrated that keccak256 hashing algorithm and secp256k1 elliptic curve signatures ensure transactions’ maximum protection, reliability, and non-rollability. The developed REV technique based on Ethereum blockchain platform increases the efficiency of data security and confidentiality, transparency and anonymity of the voting procedure, and solves the problem of coercion. The results of the work have been implemented programmatically and can be used not only in the electoral system, but also wherever there is need of remote voting.

A Secure and Efficient Multi-Factor Authentication Algorithm for Mobile Money Applications

With the expansion of smartphone and financial technologies (FinTech), mobile money emerged to improve financial inclusion in many developing nations. The majority of the mobile money schemes used in these nations implement two-factor authentication (2FA) as the only means of verifying mobile money users. These 2FA schemes are vulnerable to numerous security attacks because they only use a personal identification number (PIN) and subscriber identity module (SIM). This study aims to develop a secure and efficient multi-factor authentication algorithm for mobile money applications. It uses a novel approach combining PIN, a one-time password (OTP), and a biometric fingerprint to enforce extra security during mobile money authentication. It also uses a biometric fingerprint and quick response (QR) code to confirm mobile money withdrawal. The security of the PIN and OTP is enforced by using secure hashing algorithm-256 (SHA-256), a biometric fingerprint by Fast IDentity Online (FIDO) that uses a standard public key cryptography technique (RSA), and Fernet encryption to secure a QR code and the records in the databases. The evolutionary prototyping model was adopted when developing the native mobile money application prototypes to prove that the algorithm is feasible and provides a higher degree of security. The developed applications were tested, and a detailed security analysis was conducted. The results show that the proposed algorithm is secure, efficient, and highly effective against the various threat models. It also offers secure and efficient authentication and ensures data confidentiality, integrity, non-repudiation, user anonymity, and privacy. The performance analysis indicates that it achieves better overall performance compared with the existing mobile money systems.

ECDSA: The Virtual Signature Set of Rules of a Higher Internet

Computerized Currency is an electronic kind of cash. These days, everything is developing into digitization measure. This contains all properties like actual cash and furthermore permits prompt trades which will be reliably executed across the world while partner with upheld contraptions and organizations. In this paper we presented the SHA3-512 bit hashing algorithm and ECDSA algorithm for generation of digital signature. The Elliptic curve cryptography (ECC) is one of the greater promising technology on this area. ECC-enabled TLS is quicker and greater scalable on our servers and presents the equal or higher protection than the default cryptography in use at the web. one of the elliptic curve algorithm, the elliptic curve virtual signature algorithm (ECDSA), may be used to enhance overall performance at the Internet. CloudFlare now helps custom ECDSA certificate for our clients and that’s true for all people the use of the Internet.

Development of a hash algorithm based on cellular automata and chaos theory

Information security, reliability of data transfer are today an important component of the globalization of information technology. Therefore, the proposed work is devoted to highlighting the results of the design and development of a hacking-resistant algorithm to ensure the integrity of information transfer via digital technology and computer engineering. To solve such problems, cryptographic hashing functions are used. In particular, elements of deterministic Chaos were introduced into the developed cyclic hashing algorithm. The investigation analyzes in detail the strengths and weaknesses of known hashing algorithms. They are shown to have disadvantages. The main ones are a large number of matches (Hamming (x, y) and the presence of a weak avalanche effect, which lead to a significant decrease in the reliability of the algorithm for hacking. The designed hashing algorithm uses an iterative Merkley-Damgard structure, augmented by the input message to a length multiple of 512 bits. Processing in blocks of 128-bit uses cellular automata with mixed rules of 30, 105 and 90, 150 and takes into account the dependence of the generation of the initial vector on the incoming message. This allows half of the 10,000 pairs of arbitrary messages to have an inverse Hamming distance of 0 to 2. The proposed algorithm is four times slower than the well-known family of "secure hash algorithms." However, computation speed is not a critical requirement for a hash function. Decreasing the sensitivity to the avalanche effect allows the generation time to be approximately halved. Optimization of the algorithm, as well as its testing was carried out using new technologies of the Java programming language (version 15). Suggestions and recommendations for improving this approach to data hashing are given also