Multiple Modules Research Articles

IMD-Net: Interpretable multi-scale detection network for infrared dim and small objects.

This study proposed an interpretable multi-scale infrared small object detection network (IMD-Net) design method to improve the precision of infrared small object detection and contour segmentation in complex backgrounds. To this end, a multi-scale object enhancement module was constructed, which converted artificially designed features into network structures. The network structure was used to enhance actual objects and extract shallow detail and deep semantic features of images. Next, a global object response, channel attention, and multilayer feature fusion modules were introduced, combining context and channel information and aggregated information, selected data, and decoded objects. Finally, the multiple loss constraint module was constructed, which effectively constrained the network output using multiple losses and solved the problems of high false alarms and high missed detections. Experimental results showed that the proposed network model outperformed local energy factor (LEF), self-regularized weighted sparse model (SRWS), asymmetric contextual modulation (ACM), and other state of the art methods in the intersection-over-union (IoU) and Fmeasure values by 10.8% and 11.3%, respectively. The proposed method performed best on the currently available datasets, achieving accurate detection and effective segmentation of dim and small objects in various infrared complex background images.

Illumination homogenization of highly coherent light source based on phase modulation

When using a fly’s eye lens system to homogenize highly coherent light sources, the interference effect between the sub-beams can cause a periodic speckle distribution of illumination intensity, thereby disrupting illumination uniformity. It has been shown that using a rotating optical phase-shift plate behind the fly’s eye lens can eliminate interference patterns, but it only demonstrates engineering realizations. And the theoretical analysis and technical guidance on the phase modulation method and statistical averaging method for fly’s eye lens homogenization systems are still lacking. In this work, a simulation model of fly’s eye random phase modulation homogenization system is developed and studied in detail. Each sub-beam of the fly’s eye lens is randomly phase-modulated to break the coherence condition, and the illumination intensity of multiple independent modulations is accumulated to eliminate the interference pattern. The more times the intensity is accumulated, the better the homogenization is. Meanwhile, studied in this paper are the influence of the diffraction effect on homogenization, and the influence of the sub-lens size and focal length on the homogenization, which result in the diffracting-type system and the imaging-type system respectively. For an imaging type system, it is necessary to ensure that the first fly’s eye lens is in the front focal plane of the second fly’s eye lens. By optimizing the parameters of the fly’s eye lens and using an imaging-type system with <i>p</i> = 1.8 mm and <i>f</i><sub>A</sub> = 9 mm, a Gaussian beam with the non-uniformity of 117% is homogenized into a flat-topped beam with the non-uniformity of 1.2% in a square illumination area of 100 mm<sup>2</sup>. This fly’s eye lens random phase modulation homogenization system has a simple structure, low energy loss, and good illumination uniformity, and can be used in systems that require high coherent laser input and high resolution. This technology can be used in the field of deep-ultraviolet mask defect detection.

A novel CBAMs-BiLSTM model for Chinese stock market forecasting

The convolutional block attention module (CBAM) has demonstrated its superiority in various prediction problems, as it effectively enhances the prediction accuracy of deep learning models. However, there has been limited research testing the effectiveness of CBAM in predicting stock indexes. To address this gap and improve the prediction accuracy of stock indexes, we propose a novel model called CBAMs-BiLSTM, which combines multiple CBAM modules with a bidirectional long short-term memory network (BiLSTM). In this study, we employ the standard metric evaluation method (SME) and the model confidence set test method (MCS) to comprehensively evaluate the superiority and robustness of our models. We utilize several representative Chinese stock index data sets, namely, the Shanghai (Securities) Composite Index and the Shenzhen Composite Index, as our experimental data. The numerical results demonstrate that CBAMs-BiLSTM outperforms BiLSTM alone, achieving average reductions of 13.06%, 13.39%, and 12.48% in MAE, RMSE, and MAPE, respectively. These findings confirm that CBAM can effectively enhance the prediction accuracy of BiLSTM. Furthermore, we compare our proposed model with other popular models and examine the impact of changing data sets, prediction methods, and the size of the training set. The results consistently demonstrate the superiority and robustness of our proposed model in terms of prediction accuracy and investment returns.

Single-cell Heterogeneity-aware Transformer-guided Multiple Instance Learning for Cancer Aneuploidy Prediction from Whole Slide Histopathology Images.

Aneuploidy is a hallmark of aggressive malignancies associated with therapeutic resistance and poor survival. Measuring aneuploidy requires expensive specialized techniques that are not clinically applicable. Deep learning analysis of routine histopathology slides has revealed associations with genetic mutations. However, existing studies focus on image patches or tiles, and there is no prior work that predicts aneuploidy using single-cell analysis. Here, we present a single-cell heterogeneity-aware and transformer-guided deep learning framework to predict aneuploidy from whole slide histopathology images. First, we perform nuclei segmentation and classification to obtain individual cancer cells, which are clustered into multiple subtypes. The cell subtype distributions are computed to measure cancer cell heterogeneity. Additionally, morphological features of different cell subtypes are extracted. Further, we leverage a multiple instance learning module with Transformer, which encourages the network to focus on the most informative cancer cells. Lastly, a hybrid network is built to unify cell heterogeneity, morphology, and deep features for aneuploidy prediction. We train and validate our method on two public datasets from TCGA: lung adenocarcinoma (LUAD) and head and neck squamous cell carcinoma (HNSC), with 339 and 245 patients. Our model achieves promising performance with AUC of 0.818 (95% CI: 0.718-0.919) and 0.827 (95% CI: 0.704-0.949) on the LUAD and HNSC test sets, respectively. Through extensive ablation and comparison studies, we demonstrate the effectiveness of each component of the model and superior performance over alternative networks. In conclusion, we present a novel deep learning approach to predict aneuploidy from histopathology images, which could inform personalized cancer treatment.

A Current and Output Power Balance Method Applied in IPT System with Multiple Parallel Pickup Modules for Rail Transit

A Current and Output Power Balance Method Applied in IPT System with Multiple Parallel Pickup Modules for Rail Transit

Fusion-Embedding Siamese Network for Light Field Salient Object Detection

Light field salient object detection (SOD) has shown remarkable success and gained considerable attention from the computer vision community. Existing methods usually employ a single-/two-stream network to detect saliency. However, these methods can only handle up to two different modalities at a time, preventing them from being able to fully explore the rich information in multi-modal light field derived data. To address this, we propose the first joint multi-modal learning framework, called FES-Net, for light field SOD, which can take rich inputs not limited to two modalities. Specifically, we propose an attention-aware adaptation module to first transform the multi-modal inputs for use in our joint learning framework. The transformed inputs are then fed to a Siamese network along with multiple embedded feature fusion modules to extract informative multi-modal features. Finally, we predict saliency maps from the high-level extracted features using a saliency decoder module. Our joint multi-modal learning framework effectively resolves the limitations of existing methods, providing efficient and effective multi-modal learning that can fully explore the valuable information in light field data for accurate saliency detection. Furthermore, we improve the performance by introducing the Transformer as our backbone network. To the best of our knowledge, the improved version of our model, called FES-Trans, is the first attempt to address the challenging light field SOD with the powerful Transformer technique. Extensive experiments on benchmark datasets demonstrate that our models are superior light field SOD approaches and outperform cutting-edge models remarkably.

UCM-Net: A U-Net-Like Tampered-Region-Related Framework for Copy-Move Forgery Detection

Copy-move forgery causes a big challenge to copy-move forgery detection (CMFD) due to that the photometrical characteristics of genuine and tampered regions in the same image remain highly consistent. A novel U-Net-like architecture with multiple asymmetric cross-layer connections associated with self-correlation and atrous spatial pyramid pooling (ASPP) between feature extraction module (FEM) and tampered region localization module (TRLM), called UCM-Net, is proposed in this paper. Different from existing deep learning based CMFD networks which indiscriminately process large or small tampered regions without considering the statistical characteristics of regions, FEM differentially treats large or small tampered regions by exploiting deep backbone networks to extract high-level features with rich semantic information for large tampered regions while utilizing lightweight backbone networks to extract low-level features for small tampered regions. Multiple cross-layer connections between two modules utilize the self-correlation calculation and ASPP to remove as much irrelevant semantic information as possible while retaining multi-scale tampered features from shallow to deep convolutional layers of FEM. Unlike the previous CMFD networks, which cannot capture multi-scale features because of simply stacking convolution blocks in the upsampling step, TRLM exploits multiple U-shaped residual U-block modules with different depths to change the receptive field of each point in the tampered feature maps so as to capture global and local information, greatly improving the localization accuracy of tampered regions. Experimental results on three publicly available databases demonstrate that UCM-Net outperforms several state-of-the-art algorithms in terms of various evaluation metrics.

A Hybrid Modulation Technique for Operating Medium-Voltage High-Power CHB Converters Under Grid Voltage Disturbances

Power converters, and particularly multilevel converters, are key elements in the current energy scenario. Among them, Cascaded H-Bridge Converters (CHB) are a popular and versatile solution for many industrial applications. Technical literature provides multiple modulation techniques for its operation. Among them, hybrid-PWM modulation techniques provide a good trade-off between switching loss and output voltage quality. In this work a hybrid modulation technique is presented which allows not only the fundamental block switching operation of some power cells but also a harmonic control over a specific harmonic distortion present in the grid. To validate the operation of the proposal, several experimental results have been reported to demonstrated the good behaviour of the proposed modulation technique.

L-Fuzzy Hollow Modules and L-Fuzzy Multiplication Modules

In this paper, we give some characterizations of L-fuzzy hollow modules and of L-fuzzy multiplication modules.

Performance Analysis of a Novel Multiple Access Spatial Modulation Architecture based on STBC-OFDM and CDMA over Nakagami-m fading Channel

One of the schemes which are likely to be adopted in the forthcoming wireless communication standards is the so-called spatial modulation (SM). However, the wireless channels pertaining to such standards will mostly be frequency selective because of the involved high data rates, and SM technique alone can not overcome the resulting signal distortion. In this context, orthogonal frequency division multiplexing (OFDM) has been used as an efficient solution to combat this impediment, and space-time block coding (STBC)-OFDM has followed as an innovative approach merging the viable features of OFDM and the reliability enhancement capabilities of STBC. Obviously, SM solutions in the presence of multiple users communicating in frequency selective channels should be targeted. To fulfill such a need, a new SM-based architecture incorporating STBC-OFDM and CDMA is proposed in this work and is shown to exhibit a performance which is superior to state-of-the-art alternatives. Furthermore, its sensitivity to the aforementioned adverse parameters is studied in Nagakami channel.

Multidisciplinary design and optimization of intelligent Distributed Satellite Systems for EARTH observation

Recent advances in small, connected and intelligent satellite systems have created a wide range of opportunities for the adoption of intelligent Distributed Satellite Systems (iDSS) in communication, navigation and Earth Observation (EO) missions. iDSS are goal-oriented systems comprising of multiple satellites or modules that interact, communicate and/or cooperate with each other to accomplish the desired mission goals. The ability to mass-produce low-cost small satellites and contemporary developments in avionics/astrionics technology have spurred interest in iDSS, especially for Low Earth Orbit (LEO) satellite constellations and regional clusters. The SmartSat Cooperative Research Centre (CRC) and Australian space roadmap, as well as the landmark National Space Programme strategy and priorities, encompass EO. To date, insufficient progress and no conclusive outcome was made in terms of how contemporary Multidisciplinary Design Optimization (MDO) models and tools can be best tailored to the new capabilities and specificities of iDSS. The MDO of iDSS is challenging because it introduces new variables and highly non-linear interactions. In this context, we propose an MDO methodology to optimize an iDSS for persistent coverage over the entire Australian landmass. Several aspects of the iDSS are considered in this work, including the constellation model, subsystem models and the coupling interactions between different satellite subsystems and constellation design parameters. The constellation configuration, as well as the subsystems, are modelled using OpenMDAO, which is used to analyze and visualize the planned iDSS EO mission. The iDSS is then optimized using the Multidisciplinary Feasible (MDF) architecture approach and the iDSS interdependencies are numerically treated using the Nonlinear Block Gauss-Seidel (NLBGS) iterative solver. The resulting N2 diagrams are presented and the proposed solution is both spatially and temporally optimized, demonstrating that the proposed iDSS will enable near real-time persistent coverage over the entire Australian continent.

Design of a Configurable Five-Stage Pipeline Processor Core Based on RV32IM

With the rapid development of the electronics industry, the scale of the global Internet of Things (IoT) industry has shown an exponential growth trend in recent years. The huge demand for IoT equipment makes low cost an important indicator for the sustainable operation of the entire IoT system. However, IoT chips also require a certain amount of performance to perform complex tasks. Aiming at the above contradiction between performance and cost, this paper proposes a configurable five-stage pipeline processor core based on RV32IM. The proposed processor core has multiple configurable modules to suit different application scenarios. In low-power mode, the proposed architecture implements only an RV32I subset, while in high-performance mode, integer division and multiplication extensions are added. Meanwhile, the processor core will also support super and user privilege levels and is equipped with CSR (Control and Status Registers). The module-level and system-level simulations of the proposed architecture are completed using a fully open-source workflow based on verilator and gtkwave. In addition, the design was prototyped and verified with FPGA. The proposed processor outperforms the performance of the classic MCU-CortexM3.

State-of-Charge Balancing Control for Dual-Bus Battery System with Low-Voltage Output Regulation

This article introduces a new method for balancing the state of charge (SOC) in a dual-bus battery system architecture. The system consists of multiple battery cells or modules connected in series to provide high voltage output. Additionally, low-power flyback converters are connected in series with each battery cell or module at the inputs, and their outputs are connected in parallel to provide lower voltage output. The SOC balancing algorithm ensures that the lower voltage output remains at a desired reference value by adjusting the average duty cycle of each power converter, while also balancing the rate of charge or discharge of each battery cell or module. This SOC balancing process does not affect the normal operation of the high voltage power output. In other words, the dual output (high voltage and low voltage) of the battery system can function independently, and the balancing current only flows through the low voltage power path. Experimental results from a prototype are provided and discussed to validate the proposed dual-bus battery system and controller.

Optimization of the visual communication path of data news in the context of the Internet

Abstract Visual news belongs to a kind of data news, which uses emerging technology to bring users a new reading experience and improve the news dissemination effect. In this paper, we utilize the self-attention mechanism and BiLSTM model to extract the content of data news documents and select and classify the audio features of data news by the AdaBoost algorithm. ResNet50 is used as the backbone network, combined with multiple residual unit modules for news multi-scale image feature extraction, and the text, audio and image features of the data news are fused by constructing a multimodal feature adaptive fusion model to further optimize the visual display of the data news. This paper verifies the model’s effectiveness for the data news visualization technique through multiple perspectives, including training loss curve, performance comparison, and ablation experiment. The results show that the loss value of the model based on the Self-Attention-BiLSTM model for news text feature extraction is 0.211 after 150 iterations. The AdaBoost-based news audio feature classification method has an average classification accuracy of 97.56% for the positive diagonal when the noise is raised from 0dB to 30 dB. By adding the residual unit module to the ResNet50 backbone network, the accuracy of the multi-scale image feature extraction model has been improved by 2.77% compared to the single backbone network. The information age requires the full integration of technology and news and combining the corresponding Internet platform for the visual display of data news in order to promote the expansion of the visual communication path of data news.

Dynamic multifunctional metasurfaces: an inverse design deep learning approach

Optical metasurfaces (OMs) offer unprecedented control over electromagnetic waves, enabling advanced optical multiplexing. The emergence of deep learning has opened new avenues for designing OMs. However, existing deep learning methods for OMs primarily focus on forward design, which limits their design capabilities, lacks global optimization, and relies on prior knowledge. Additionally, most OMs are static, with fixed functionalities once processed. To overcome these limitations, we propose an inverse design deep learning method for dynamic OMs. Our approach comprises a forward prediction network and an inverse retrieval network. The forward prediction network establishes a mapping between meta-unit structure parameters and reflectance spectra. The inverse retrieval network generates a library of meta-unit structure parameters based on target requirements, enabling end-to-end design of OMs. By incorporating the dynamic tunability of the phase change material Sb2Te3 with inverse design deep learning, we achieve the design and verification of dynamic multifunctional OMs. Our results demonstrate OMs with multiple information channels and encryption capabilities that can realize multiple physical field optical modulation functions. When Sb2Te3 is in the amorphous state, near-field nano-printing based on meta-unit amplitude modulation is achieved for X-polarized incident light, while holographic imaging based on meta-unit phase modulation is realized for circularly polarized light. In the crystalline state, the encrypted information remains secure even with the correct polarization input, achieving double encryption. This research points towards ultra-compact, high-capacity, and highly secure information storage approaches.

Application of Different Python Libraries for Visualisation of Female Genital Mutilation

Utilizing data visualization facilitates the analysis and comprehension of common data provided by the media, individuals, governments, and other sectors. Python is a well-known programming language that excels at scientific data visualization. This thesis utilizes a variety of Python modules, including Pandas, NumPy, Matplotlib, Seaborn, Plotly, and Bokeh, to illustrate female genital mutilation. The purpose of this thesis is to illustrate female genital mutilation and explain its performance pattern using a complex, interactive diagram that integrates multiple types of Python libraries. In comparison to other libraries, Plotly is the simplest, yet it performs at the highest level. NumPy and Matplotlib are combined to produce Hexbins charts. NumPy provides an N-dimensional plot, and Matplotlib allows for the plot's colours to be customized. Despite its limited customization options, the Seaborn library is suitable for both data visualization and statistical modelling. Due to this deficiency, the Seaborn library is frequently combined with Matplotlib to generate superior visualizations. As a result, this thesis will be recommended to both specialists and novices as worthwhile reading. In addition, it will assist the government in drafting legislation to end female genital mutilation. They will comprehend the significance of combining multiple Python modules to generate intricate interactive diagrams for data visualization in the field of data science. This information will be posted online to contribute to the corpus of knowledge.

Concerning the Nakayama property of a module

Abstract In this paper, we thoroughly study the Nakayama property and some related concepts. Also, we describe multiplication modules that, among other things, satisfy the Nakayama property. Next, we show that a ring 𝑅 is a Max ring if and only if all modules that can be generated by a finite or countable set have the weak Nakayama property. We prove that a ring 𝑅 is a perfect ring if and only if every module that can be generated by a finite or countable set has the Nakayama property. Finally, we present some categorical results on the aforementioned properties.

Development of supporting platform for the fine flow characteristics of reactor core

This study presents the Supporting platform for reactor fine flow characteristics calculation and analysis (Cilian platform), a user-friendly tool that supports the analysis and optimization of pressurized water reactor (PWR) cores with mixing vanes using computational fluid dynamics (CFD) computing. The Cilian platform allows for easy creation and optimization of PWR's main CFD calculation schemes and autonomously manages CFD calculation and analysis of PWR cores, reducing the need for human and computational resources. The platform's key features enable efficient simulation, rapid solution design, automatic calculation of core scheme options, and streamlined data extraction and processing techniques. The Cilian platform's capability to call external CFD software reduces the development time and cost while improving the accuracy and reliability of the results. In conclusion, the Cilian platform exemplifies an innovative solution for efficient computational fluid dynamics analysis of pressurized water reactor (PWR) cores. It holds great promise for driving advancements in nuclear power technology, enhancing the safety, efficiency, and cost-effectiveness of nuclear reactors. The platform adopts a modular design methodology, enabling the swift and accurate computation and analysis of diverse flow regions within core components. This design approach facilitates the seamless integration of multiple computational modules across various reactor types, providing a high degree of flexibility and reusability.

Learnable PM diffusion coefficients and reformative coordinate attention network for low dose CT denoising

Objective. Various deep learning methods have recently been used for low dose CT (LDCT) denoising. Aggressive denoising may destroy the edge and fine anatomical structures of CT images. Therefore a key issue in LDCT denoising tasks is the difficulty of balancing noise/artifact suppression and edge/structure preservation. Approach. We proposed an LDCT denoising network based on the encoder-decoder structure, namely the Learnable PM diffusion coefficient and efficient attention network (PMA-Net). First, using the powerful feature modeling capability of partial differential equations, we constructed a multiple learnable edge module to generate precise edge information, incorporating the anisotropic image processing idea of Perona–Malik (PM) model into the neural network. Second, a multiscale reformative coordinate attention module was designed to extract multiscale information. Non-overlapping dilated convolution capturing abundant contextual content was combined with coordinate attention which could embed the spatial location information of important features into the channel attention map. Finally, we imposed additional constraints on the edge information using edge-enhanced multiscale perceptual loss to avoid structure loss and over-smoothing. Main results. Experiments are conducted on simulated and real datasets. The quantitative and qualitative results show that the proposed method has better performance in suppressing noise/artifacts and preserving edges/structures. Significance. This work proposes a novel edge feature extraction method that unfolds partial differential equation into neural networks, which contributes to the interpretability and clinical application value of neural network.

Cyclic Heat Transfer Solver for OpenFOAM

Channels with periodically repeating geometries are often simulated using periodic or cyclic boundary conditions. By calculating the temperature and flow field in one periodic module, the resulting distributions can be generalized to multiple modules. This reduces the computational load by simulating a single module versus the whole structure. This is a particularly useful approach when performing large optimisation studies of periodic geometries, such as compact heat exchangers. Currently, OpenFOAM only supports cyclic boundary conditions for pressure and momentum, but not heat transfer. The present work introduces a steady and an unsteady solver for cyclic heat transfer with constant wall temperature boundary conditions. The solver is validated against analytical Hagen-Poiseuille flow and two configurations of periodic wavy channels. In the latter case, the results are compared to existing literature.