Basic Transformations Research Articles

Adaptive Spatial Transformation Networks for Periocular Recognition

Periocular recognition has emerged as a particularly valuable biometric identification method in challenging scenarios, such as partially occluded faces due to COVID-19 protective masks masks, in which face recognition might not be applicable. This work presents a periocular recognition framework based on deep learning, which automatically localises and analyses the most important areas in the periocular region. The main idea is to derive several parallel local branches from a neural network architecture, which in a semi-supervised manner learn the most discriminative areas in the feature map and solve the identification problem solely upon the corresponding cues. Here, each local branch learns a transformation matrix that allows for basic geometrical transformations (cropping and scaling), which is used to select a region of interest in the feature map, further analysed by a set of shared convolutional layers. Finally, the information extracted by the local branches and the main global branch are fused together for recognition. The experiments carried out on the challenging UBIRIS-v2 benchmark show that by integrating the proposed framework with various ResNet architectures, we consistently obtain an improvement in mAP of more than 4% over the "vanilla" architecture. In addition, extensive ablation studies were performed to better understand the behavior of the network and how the spatial transformation and the local branches influence the overall performance of the model. The proposed method can be easily adapted to other computer vision problems, which is also regarded as one of its strengths.

AOP-Based Transformation of Abacavir in Different Environments: Evolution Profile of Descyclopropyl-Abacavir and In Silico Toxicity Assessment of the Main Transformation Products

This study explores the photocatalytic transformation of the antiviral drug abacavir employing different advanced oxidation processes (AOPs) such as UV/TiO2, UV/MOF/H2O2, UV/MOF/S2O82-, UV/Fe2+/H2O2, and UV/Fe2+/S2O82-. All processes appear to be effective in eliminating abacavir within a few minutes, while the evolution profile of the basic transformation product, descyclopropyl-abacavir (TP-247) was also monitored. Moreover, the implementation of the most efficient technologies towards the removal of abacavir in different matrices such as wastewater effluent and leachate was also assessed, revealing that the organic matter present or the inorganic constituents can retard the whole process. Four major transformation products were detected, and their time-evolution profiles were recorded in all studied matrices, revealing that different transformation pathways dominate in each matrix. Finally, the prediction of the toxicity of the major TPs employing ECOSAR software was conducted and showed that only hydroxylation can play a detoxification role in the treated solution.

가려진 사람의 재식별을 위한 트랜스포머 기반 교차 어텐션과 특징 다양화

Occluded person re-identification is a very difficult because the specific person is occluded by obstacles or other persons or by oneself. Major works adopt transformer-based approach show excellent performances, but they used a basic transformer only. In this paper, we suggest the various techniques to improve the transformer-based Re-ID method for the occluded person as follows. First, after extracting the heatmap and then deleting random body parts on the heat map, accurate keypoint information is obtained in data augmentation. Second, Cross-attention between the keypoint heatmap and the output of the transformer

Finite-time boundedness and finite-time stabilization boundedness for nonlinear interconnected systems with Atangana–Baleanu–Caputo fractional derivative

In this paper, we consider the finite-time boundedness (FTB) and finite-time stabilization boundedness (FTSB) problems for a class of nonlinear interconnected systems with Atangana–Baleanu–Caputo (ABC) fractional derivative. By employing the FTB theory and some properties of ABC fractional derivative, and some basic mathematical transformations, we derive new conditions for the FTB and FTSB problems of ABC fractional-order nonlinear interconnected systems. These conditions are formed in linear matrix inequalities (LMIs), which can be effectively solved using the MATLAB LMI Control Toolbox. Finally, two numerical examples with simulations are provided to illustrate the efficiency of the proposed method.

PREDICTIVE ASSESSMENT OF DIESEL FUEL QUALITY BASED ON WAVELET TRANSFORMATIONS

The principle of device operation for determining the quality control of diesel fuel is considered, which helps to investigate the spectral characteristics of diesel fuel. Basic wavelet transformations are given. IR spectroscopy of diesel fuel from different producing countries was carried out, and determination of the cetane number through wavelet coefficients and determining conclusion diesel fuel quality were performed.

Cascade Transformer Decoder Based Occluded Pedestrian Detection With Dynamic Deformable Convolution and Gaussian Projection Channel Attention Mechanism

Occluded pedestrian detection is very challenging in computer vision, because the pedestrians are frequently occluded by various obstacles or persons, especially in crowded scenarios. In this paper, an occluded pedestrian detection method is proposed under a basic DEtection TRansformer (DETR) framework. Firstly, Dynamic Deformable Convolution (DyDC) and Gaussian Projection Channel Attention (GPCA) mechanism are proposed and embedded into the low layer and high layer of ResNet50 respectively, to improve the representation capability of features. Secondly, Cascade Transformer Decoder (CTD) is proposed, which aims to generate high-score queries, avoiding the influence of low-score queries in the decoder stage, further improving the detection accuracy. The proposed method is verified on three challenging datasets, namely CrowdHuman, WiderPerson, and TJU-DHD-pedestrian. The experimental results show that, compared with the state-of-the-art methods, it can obtain a superior detection performance.

ПЕРЕВОД РОМАНА ФИЛИПА К. ДИКА “DO ANDROIDS DREAM OF ELECTRIC SHEEP?”: НЕОБОСНОВАННОСТЬ ТРАНСФОРМАЦИЙ И СМЫСЛОВЫЕ ОШИБКИ

As the result of comparative analysis performed on the original text of the novel and its translation into Russian, several fragments were revealed where the translator made semantic mistakes, some of them being quite rough. The study carried out to identify the reasons of this phenomenon to occur shows that some of these mistakes were caused by the unreasonable use of lexical-grammatical transformations during the translation of the text, primarily the transformations of addition and omission. Other semantic mistakes appeared in the translation either because the translator was not paying enough attention to what he was doing or he failed to read the context properly. The paper provides the brief review of basic lexical-grammatical transformations used in the process of translation, as well as the analysis of translation of several fragments in the novel containing rough semantic mistakes. The results of the research are used to draw some conclusions regarding the causes for the semantic mistakes to occur; the paper also figures out the role of the unreasonable lexical-grammatical transformations in making some of these mistakes.

A Digital Transforming Plan for Vehicle Industry: Evidence from Audi

With the development of the Internet and the impact of COVID-19, many vehicle companies have started digital transformation in order to unlock the potential of the sinking market and promote profits by developing online service. The direct-sales model and customization are highly compatible with the online service and provides a better development environment for new energy vehicles. Based on the strong economic foundation of FAW Audi, this study investigates the building of a digital system to contact customers online of it. To start the direct-sales model, website, mini-programs, and applications will connect customers with the company rather than 4S stores. Collecting and analyzing customers’ needs from basic automobile elements (e.g., color and shape) to each component through these online methods to start customized businesses, so as to achieve basic digital transformation and change the relatively backward digital level in the Chinese vehicle market. In addition, through the digital system, the company, driverless vehicles and customers are connected together which is the basic of starting online car hailing business, by which FAW Audi will further enhance its influence. These results shed light on guiding further exploration of digital transforming in vehicle industry.

Reading is not believing: A multimodal adversarial attacker for Chinese-NLP model

The research of adversarial examples has extended from image to text in the last few years. However, these attacks are typically limited to the English language and simple substitution strategies. To further expose the vulnerability of NLP models, we study the linguistic characteristics of Chinese, the quintessential ideogram with over 1.2 billion native speakers. Accordingly, a novel attack framework named ZH-Deceiver is proposed to generate Chinese adversarial examples from the perspective of morphology, phonetics, semantics, and basic transformation. In particular, a CNN-based Siamese Network is integrated to ameliorate the quality of adversarial examples. To elaborate the validity of ZH-Deceiver, extensive experiments are conducted on two datasets. Compared with four benchmarks such as Genetic, PWWS, TextBugger, and SememePSO, our attack achieves impressive performance on effectiveness, efficiency, imperceptibility, and human evaluation by deceiving seven AI models including CNN and BERT. Furthermore, the transferability, as well as the robustness, is further analyzed and the former is successfully applied to attack three commercial APIs: Tencent, ALi, and Baidu. ZH-Deceiver acts as a wake-up call for multilingual processing models, and tangibly extends the application and methodology of adversarial textual attack.

Ligand enabled none-oxidative decarbonylation of aliphatic aldehydes

Decarbonylation of aldehydes is a basic organic transformation, which has been developed for more than six-decade. However, as comparing to well-studied aromatic aldehydes, fewer examples for catalytic decarbonylation of aliphatic aldehydes were reported, mainly on simple or special substrates. For α-bulky or highly functionalized ones, stoichiometric Rh(I) were usually required for decent yields. Herein, we present a rare example of Ir(I)-catalyzed direct decarbonylation of α-quaternary aldehydes with broad substrate scope and good functional group compatibility via judicious selection of ligand. The α-chirality is memorized in this decarbonylation process. In addition, we report a broad-spectrum decarbonylation of α-secondary and α-tertiary aldehydes containing multifunctional groups with an improved Rh(I)/DPPP recipe. Finally, we realized selective decarbonylation of α-tertiary aldehydes in the presence of α-quaternary one via the reactivity differences.

Solution Existence Theory for Artificial Neural Networks

Artificial neural networks (NNs) have been widely used in creation of various machine learning (ML) models. Training an NN for a given dataset is essentially finding a set of weights and biases, known as learning (or training) parameters. An NN uses the affine transformations to encode the features in the dataset in the form of learning parameters. One may naturally ask fundamental questions: (1) can the NN be trained, meaning that can a solution for a set of learning parameters can be found? and (2) is the solution unique? This paper addresses these questions. First, detailed formulation is presented for a simple NN with just one basic affine transformation unit (ATU) for linear regression problems. Conditions for solution existence and uniqueness will be examined, leading to the Solution Existence Theory. A discussion will be presented when this Solution Existence Theory is extended to general NNs for complex problems, in addition to a discussion on predictability of NNs.

The Software Cache Optimization-Based Method for Decreasing Energy Consumption of Computational Clusters

Reducing the consumption of electricity by computing devices is currently an urgent task. Moreover, if earlier this problem belonged to the competence of hardware developers and the design of more cost-effective equipment, then more recently there has been an increased interest in this issue on the part of software developers. The issues of these studies are extensive. From energy efficiency issues of various programming languages to the development of energy-saving software for smartphones and other gadgets. However, to the best of our knowledge, no study has reported an analysis of the impact of cache optimizations on computing devices’ power consumption. Hence, this paper aims to provide an analysis of such impact on the software energy efficiency using the original software design procedure and computational experiments. The proposed Software Cache Optimization (SCO)-based Methodology was applied to one of the key linear algebra transformations. Experiments were carried out to determine software energy efficiency. RAPL (Running Average Power Limit) was used—an interface developed by Intel, which provides built-in counters of Central Processing Unit (CPU) energy consumption. Measurements have shown that optimized software versions reduce power consumption up to 4 times in relation to the basic transformation scheme. Experimental results confirm the effectiveness of the SCO-based Methodology used to reduce energy consumption and the applicability of this technique for software optimization.

Visual Anthropology: Formation Stages and Basic Elements of Analysis

The article contributes to the history systematization of the visual anthropology area. The author considers and conceptualizes the stages of this discipline formation not only in accordance with the logic of self-understanding of its representatives, but also taking into account the genesis of optical media. The parameters of video production prove not only the instrumental role of visual anthropology in relation to the field of cultural anthropology, but also allow the latter to be a science in the strict sense of the term; that is, to have not only theoretical generalizations but also a rich empirical base. The inability of textual forms of recording anthropological material to adequately capture the cultural practices of traditional communities has also revealed the preserving and even salvage potential of the video production. However, the dominant of writing as a basic practice of science and its definition as a transparent carrier of scientific discourse did not allow to understand, at the initial stages. the innovative potential of visual anthropology, the specifics of its optics and methodology. The article pays attention both to the specifics of the practice of fixing video products (painting, photography, cinema, and the Internet) and to the forms of the representatives reflection of anthropological thought about their influence on the anthropology subject field. In this article, particular attention is paid to the degree of differentiation of cultural anthropology subject fields and visual anthropology against the background of basic transformations of cultural research. The influence of basic theoretical guidelines, in particular the principle of historical rationalism, participation in the formation of visual anthropology area itself, is also defined. In addition to theoretical principles and procedures of description, as well as comprehension of visual products and guidelines of research communities, the methodological significance of other parameters, formed as basic units of visual anthropology, are analyzed: technical parameters of optical media, the order of signifiers of visual representations, communication between video production and the audience. The author presents the disciplinary and historical context of the genesis of visual anthropology, as well as analyzes the conceptual logic of collective work edited by Paul Hockings “Principles of Visual Anthropology,” which is considered a fundamental work for self-awareness of this research area.

Symmetry of terminating basic hypergeometric series representations of the Askey–Wilson polynomials

In this paper, we explore the symmetric nature of the terminating basic hypergeometric series representations of the Askey–Wilson polynomials and the corresponding terminating basic hypergeometric transformations that these polynomials satisfy. In particular we identify and classify the set of 4 and 7 equivalence classes of terminating balanced ϕ34 and terminating very-well-poised W78 basic hypergeometric series which are connected with the Askey–Wilson polynomials. We study the inversion properties of these equivalence classes and also identify the connection of both sets of equivalence classes with the symmetric group S6, the symmetry group of the terminating balanced ϕ34. We then use terminating balanced ϕ34 and terminating very-well poised W78 transformations to give a broader interpretation of Watson's q-analog of Whipple's theorem and its converse.

Transformational Change: Challenges and Development Prospects. The Case of Ukraine

Background: The basic system transformation approaches can be described by definition of the main and decisive factors of transformational change.
 Research objectives: The aim of the study is to estimate how the GDP in Ukraine depends on the degree of economic, political, social transformation, as well as the overall indicators (market and democratic) reforms.
 Research design and methods: The source of information is the World Bank database, Heritage Foundation, and national statistics. The research methods include descriptive analysis, economic and statistical methods, and correlation analysis.
 Results: The study results identify factors which positively impacts the formation of new and renewed institutions and contribute to a more balanced political, social and economic growth.
 Conclusions: Development vector designation allows formulating a state development strategy by estimating the factors that might lead to a more sustainable strategy of political and social-economic development.

Blind Digital Watermark Based on Discrete Fourier Transformation

The policy and security of digital images are concerned by all artists. Embedding blind digital watermarks to images is an effective solution to such problem. The existing blind digital watermark techniques for images are usually based on the spatial domain method, such as some build-in tools in some software like Photoshop. Those methods' imperceptibility is good, but the robustness is dissatisfactory. Therefore, blind digital watermark based on the frequency domain is becoming popular nowadays. The use of the frequency domain-based techniques can not only keep the imperceptibility but also ensure good robustness and non-removability. According to the reasons mentioned above, this thesis presents a frequency domain-based blind digital watermark technique. This method uses basic discrete Fourier transformation to transform an image to the frequency domain and then embeds a digital watermark. To increase the security, we also introduce some methods to encode watermarks before embedding them to the images. We conduct some experiments by applying the technique to some images with different watermarks. In the experiments, it can embed blind watermarks with the least effect on the quality of the original image and can also successfully extract watermarks from the watermarked image. The results show this technique is highly useable. We also conduct some experiments to simulate the attacks. This technique also shows great robustness after many kinds of different attacking methods.

Effects of Data Enrichment with Image Transformations on the Performance of Deep Networks

Images cannot always be expected to come in a certain standard format and orientation. Deep networks need to be trained to take into account unexpected variations in orientation or format. For this purpose, training data should be enriched to include different conditions. In this study, the effects of data enrichment on the performance of deep networks in the super resolution problem were investigated experimentally. A total of six basic image transformations were used for the enrichment procedures. In the experiments, two deep network models were trained with variants of the ILSVRC2012 dataset enriched by these six image transformation processes. Considering a single image transformation, it has been observed that the data enriched with 180 degree rotation provides the best results. The most unsuccessful result was obtained when the models were trained on the enriched data generated by the flip upside down process. Models scored highest when trained with a mix of all transformations.

Several Integrable Properties of the Discrete KP Hierarchy Obtained by Means of Gauge Transformations

The gauge transformation is a powerful method to solve integrable hierarchies. In the discrete KP hierarchy, there are two basic gauge transformations, T D ( q ) = Λ( q ) ∘ Δ ∘ q −1 and T I ( r ) = Λ −1 ( r −1 ) ∘ Δ −1 ∘ r. This paper is concerned with several integrable properties of the discrete KP hierarchy found by means of gauge transformations. The successive applications of T I to the constrained discrete KP hierarchy have been presented. Meanwhile, we derive two novel discrete equations as the orbits of gauge transformations T D and T I for the constrained KP hierarchy. Moreover, with the help of the spectral representations and the gauge transformations, Fay-similar equations of the tau functions for the constrained discrete KP have been investigated. In addition, we also discuss the relation between the gauge transformation and an additional symmetry for the discrete KP hierarchy.

Sifat Preservasi Lingkaran dan Garis Pada Transformasi Möbius

This article discusses Möbius transformation from the point of view of algebra to describe one of its geometric properties, i.e. preserving circles and lines in complex planes. In simple terms, this preservation means that Möbius transformation maps a collection of circles and lines (back) into a collection of circles and lines. In general, the discussion begins with an explanation of the definition of the Möbius transformation in the complex plane. The discussion continues on defining the basic mapping and direct affine transformation. These two concepts are used to prove the existence of the preservation properties of circles and lines in the Möbius transformation. It can be shown that the Möbius transformation can be expressed as a composition of the direct affine transform and the inverse. It can also be shown that the direct affine transform and the inverse both have the property of preserving circles and lines in the complex plane. Thus, it can be concluded that in this study the Möbius transformation has the property of preserving circles and lines in the complex plane.

Historical stages of the evolution of state regulation of banking activity in the USSR: 1920s - early 1930s

The article examines the evolution of state regulation of the banking sector in the initial period of socialist construction in the USSR. The development of the monetary system and the banking sector is considered in relation to the tasks facing the national economy of the country in the 1920s and 30s. The dynamics of these tasks determined the development of regulatory influences, the emergence of appropriate forms and mechanisms. The stages of the implementation of the credit reform of 1930s and its results are analyzed. The sequence of preparatory measures of the reform is considered, which concerned both the creation of prerequisites for the transition of enterprises to work in new monetary conditions, and the strengthening of the role of the State Bank as the main conductor of the reform. The analysis of the dynamics of the sectoral structure of lending and the changes that occurred as a result of the transition of the economy to new operating conditions. From the standpoint of a systematic approach, the structure of the system of state regulation of banking activity after the completion of the credit reform is studied. This structure is considered on the basis of the proposed organizational and functional model. The layout of the organizational and functional modeling of state banking regulation proposed by the author allows to consider the forms and mechanisms of regulation and interaction of participants in this process from the standpoint of a systematic approach. The Soviet experience of creating a system of state regulation is also interesting with a certain sequence of transformations: the preparation of preliminary measures distributed over time, then the implementation of a radical basic transformation (credit reform proper), the adjustment of decisions and practices based on the analysis of the actual results of transformations. Such an approach and aspect of the reform has been little studied in historical studies, but its understanding for modern Russian realities could be useful.