Asymmetric optical cryptosystem for multiple images based on devil’s spiral Fresnel lens phase and random spiral transform in gyrator domain

An asymmetric cryptosystem is presented for encrypting multiple images in gyrator transform domains. In the encryption approach, the devil’s spiral Fresnel lens variable pure phase mask is first designed for each image band to be encrypted by using devil’ mask, random spiral phase and Fresnel mask, respectively. Subsequently, a novel random devil’ spiral Fresnel transform in optical gyrator transform is implemented to achieved the intermediate output. Then, the intermediate data is divided into two masks by employing random modulus decomposition in the asymmetric process. Finally, a random permutation matrix is utilized to obtain the ciphertext of the intact algorithm. For the decryption approach, two divided masks (private key and ciphertext) need to be imported into the optical gyrator input plane simultaneously. Some numerical experiments are given to verify the effectiveness and capability of this asymmetric cryptosystem.

Electrical Supply Circuit for a Cold Plasma Source at Atmospheric Pressure Based on a Voltage Multiplier

A cold plasma source operating at atmospheric pressure powered by a voltage multiplier is reported. In addition to its usual high voltage output, there is an intermediate output of lower voltage and higher current capability. A discharge current is drawn from both outputs. The ratio of the current supplied by each output depends on the operating state, namely, before or after the plasma jet formation. The electrical circuit is equivalent to two dc sources connected in parallel, used to initiate and sustain the electrical discharge. The plasma source is aimed to study the effect of cold plasma on the surface of various liquid or solid materials, including polymers.

Experimental Estimates of the Student Attendance Production Function

Student attendance is both a critical input and intermediate output of the education production function. However, the malleable classroom-level determinants of student attendance are poorly understood. We estimate the causal effect of class size, class composition, and observable teacher qualifications on student attendance by leveraging the random classroom assignments made by Tennessee’s Student/Teacher Achievement Ratio (STAR) Project class size experiment. A 10-student increase in class size increases the probability of being chronically absent by about 3 percentage points (21%). For Black students, random assignment to a Black teacher reduces the probability of chronic absence by 3.1 percentage points (26%). However, naive mediation analyses suggest that attendance is not a mechanism through which class size and same-race teachers improve student achievement.

Copyful Streaming String Transducers

Copyless streaming string transducers (copyless SST) have been introduced by R. Alur and P. Černý in 2010 as a one-way deterministic automata model to define transductions of finite strings. Copyless SST extend deterministic finite state automata with a set of variables in which to store intermediate output strings, and those variables can be combined and updated all along the run, in a linear manner, i.e., no variable content can be copied on transitions. It is known that copyless SST capture exactly the class of MSO-definable string-to-string transductions, and are as expressive as deterministic two-way transducers. They enjoy good algorithmic properties. Most notably, they have decidable equivalence problem (in PSpace). On the other hand, HDT0L systems have been introduced for a while, the most prominent result being the decidability of the equivalence problem. In this paper, we propose a semantics of HDT0L systems in terms of transductions, and use it to study the class of deterministic copyful SST. Our contributions are as follows: (i)HDT0L systems and total deterministic copyful SST have the same expressive power, (ii)the equivalence problem for deterministic copyful SST and the equivalence problem for HDT0L systems are inter-reducible, in quadratic time. As a consequence, equivalence of deterministic SST is decidable, (iii)the functionality of non-deterministic copyful SST is decidable, (iv)determining whether a non-deterministic copyful SST can be transformed into an equivalent non-deterministic copyless SST is decidable in polynomial time.

G-CNN and double-referenced thresholding for detecting time series anomalies

Anomaly detection based on time series data is of great importance in many fields. Time series data produced by man-made systems usually include two parts: monitored and exogenous data, which respectively are the detected object and the control/feedback information. In this paper, a so-called G-CNN architecture that combined the gated recurrent units (GRU) with a convolutional neural network (CNN) is proposed, which respectively focus on the monitored and exogenous data. The most important is the introduction of a complementary double-referenced thresholding approach that processes prediction errors and calculates threshold, achieving balance between the minimization of false positives and the false negatives. The outstanding performance and extensive applicability of our model is demonstrated by experiments on two public datasets from aerospace and a new server machine dataset from an Internet company. It is also found that the monitored data is close associated with the exogenous data if any, and the interpretability of the G-CNN is discussed by visualizing the intermediate output of neural networks.

Performance of China's rural supply chain finance: from the perspective of maximization of intermediate output

PurposeTo dilute the financial difficulties in agricultural production and operation, the Chinese government has actively explored and developed rural supply chain finance (RSCF) service systems. The purpose of this study is to evaluate and analyze the performance of RSCF systems in China.Design/methodology/approachTo evaluate the performance of RSCF systems in China, this study proposes a two-stage data envelopment analysis model. Compared with other models, the model proposed in this study considers not only the technical gap between RSCF systems but also the maximization of intermediate output to conform to the practice of RSCF.FindingsBased on the empirical analysis, this study draws the following four conclusions. First, the overall efficiency of China's RSCF systems is low, and there remains great potential for improvement. Second, the technology gap ratio index score and meta-frontier efficiency of RSCF systems in Central China are the lowest in all regions, indicating that the technical level of RSCF systems in Central China is the lowest. Third, the relationship between rural residents' disposable income and the efficiency of RSCF systems is U-shaped, and the efficiency of RSCF systems in the high-income group is far greater than that of other income groups. Finally, the main reason for the lack of efficiency in RSCF seems to lie in management and technology.Originality/valueThis study divides all RSCF systems into four types according to management potential and technical potential, and recommend corresponding improvement suggestions for different kinds of RSCF systems.

Blood transcriptional module repertoire analysis and visualization using R

Transcriptional modules have been widely used for the analysis, visualization and interpretation of transcriptome data. We have previously described the construction and characterization of generic and reusable blood transcriptional module repertoires. The third and latest version that we have recently made available comprises 382 functionally annotated gene sets (modules) and encompasses 14,168 transcripts. We developed R scripts for performing module repertoire analyses and custom fingerprint visualization. These are made available here along with detailed descriptions. An illustrative public transcriptome dataset and corresponding intermediate output files are also included as supplementary material. Briefly, the steps involved in module repertoire analysis and visualization include: First, the annotation of the gene expression data matrix with module membership information. Second, running of statistical tests to determine for each module the proportion of its constitutive genes which are differentially expressed. Third, the results are expressed “at the module level” as percent of genes increased or decreased and plotted in a fingerprint grid format. A parallel workflow has been developed for computing module repertoire changes for individual samples rather than groups of samples. Such results are plotted in a heatmap format. The use case that is presented illustrates the steps involved in the generation of blood transcriptome repertoire fingerprints of septic patients at both group and individual levels.

Plug-in, Trainable Gate for Streamlining Arbitrary Neural Networks

Architecture optimization, which is a technique for finding an efficient neural network that meets certain requirements, generally reduces to a set of multiple-choice selection problems among alternative sub-structures or parameters. The discrete nature of the selection problem, however, makes this optimization difficult. To tackle this problem we introduce a novel concept of a trainable gate function. The trainable gate function, which confers a differentiable property to discrete-valued variables, allows us to directly optimize loss functions that include non-differentiable discrete values such as 0-1 selection. The proposed trainable gate can be applied to pruning. Pruning can be carried out simply by appending the proposed trainable gate functions to each intermediate output tensor followed by fine-tuning the overall model, using any gradient-based training methods. So the proposed method can jointly optimize the selection of the pruned channels while fine-tuning the weights of the pruned model at the same time. Our experimental results demonstrate that the proposed method efficiently optimizes arbitrary neural networks in various tasks such as image classification, style transfer, optical flow estimation, and neural machine translation.

Nonprofit Public Libraries and Organizational Performance

This study applied a two-stage data envelopment analysis (DEA) model with variable return to scale (VRS) to assess the impact of intermediate outputs on the technical efficiency of nonprofit public libraries (NPLs) in the United States (US) with respect to attaining service and program outcomes. The findings revealed that 46% of the NPLs were technically efficiency with respect to attaining the intermediate outputs at stage one. At stage two, 7% of the libraries were efficient with respect to attaining their service and program outcomes. The findings also revealed that the libraries which were efficient at stage one had an average reciprocal inefficiency score of 0.396 at stage two. By contrast, libraries which are inefficient at stage one had higher efficiency scores at stage two. The DEA analysis also produced estimates in regard to the optimal level of performance the NPLs should attain for each intermediate output to increase the level of technical efficiency at stage two.

HsgNet: A Road Extraction Network Based on Global Perception of High-Order Spatial Information

Road extraction is a unique and difficult problem in the field of semantic segmentation because roads have attributes such as slenderness, long span, complexity, and topological connectivity, etc. Therefore, we propose a novel road extraction network, abbreviated HsgNet, based on high-order spatial information global perception network using bilinear pooling. HsgNet, taking the efficient LinkNet as its basic architecture, embeds a Middle Block between the Encoder and Decoder. The Middle Block learns to preserve global-context semantic information, long-distance spatial information and relationships, and different feature channels’ information and dependencies. It is different from other road segmentation methods which lose spatial information, such as those using dilated convolution and multiscale feature fusion to record local-context semantic information. The Middle Block consists of three important steps: (1) forming a feature resource pool to gather high-order global spatial information; (2) selecting a feature weight distribution, enabling each pixel position to obtain complementary features according to its own needs; and (3) inversely mapping the intermediate output feature encoding to the size of the input image by expanding the number of channels of the intermediate output feature. We compared multiple road extraction methods on two open datasets, SpaceNet and DeepGlobe. The results show that compared to the efficient road extraction model D-LinkNet, our model has fewer parameters and better performance: we achieved higher mean intersection over union (71.1%), and the model parameters were reduced in number by about 1/4.