Split Modules Research Articles

Development of impact characteristics response model for combined tube expansion axial splitting module

Combined tube-expansion and axial splitting mechanism yields excellent force-displacement characteristic and high stroke efficiency. In this paper, a mathematical model to estimate and optimise impact response of such mechanism was developed, to correlate the impact response to the module dimensional parameters, i.e. diameter-to-thickness ratio, D1/t and expansion ratio, D2/D1 and other design parameters such as material strength, expansion angle, and friction coefficient. Accurate finite element (FE) analysis, validated with experimental results were performed to generate results sufficient to form a response surface model (RSM). The current mathematical model successfully captured the effect of expansion ratio, thickness ratio, and initial diameter of the tube, with R 2 of 0.99 for specific energy absorbed and 0.81 for mean crushing force throughout 80 data points. A case study was also presented which shows the result comparison between numerical and the proposed model, yielding error below 1% difference for mean and peak crushing force, total energy absorbed, specific energy absorbed (SEA), stroke and crushing force efficiencies. The model has also been implemented in an optimisation for railway vehicle case study using the proposed mathematical model. The current research found that the combined module could reach SEA and stroke efficiency values of 29.74 kJ/kg and 0.99, respectively, significantly better than most of the metallic-based impact energy absorbing mechanisms.

Research on Array Magnetometer Spectroscopy Technology Based on Diffractive Optical Elements

Abstract With the development of Spin-Exchange Relaxation Free (SERF) atomic magnetometers, human weak magnetic measurement has become a hot research direction in the medical field. Cardio-cerebromagnetic monitoring is a new medical detection method in the medical field. Array magnetometer is the core device of cardio-cerebromagnetic accurate monitoring. The volume and accuracy of the array magnetometer directly determine the wearability and sensitivity of the device, and the laser beam splitting system has a great influence on its accuracy and volume. Because the diffraction beam splitting method has the advantages of high beam splitting efficiency, good beam performance consistency, small space occupation, etc., it is suitable for the laser beam splitting module of array magnetometer, which does not reduce the accuracy while reducing the volume. In this paper, we innovatively apply the diffraction beam splitting scheme to the laser beam splitting module of the array magnetometer, with the goal of miniaturizing the integration of the beam splitting module and improving the output accuracy of the array magnetometer. The design and fabrication of a high-performance diffraction beam splitter applied to the beam splitting system of the SERF atomic magnetometer, as well as the output beam quality and power balance of multiple laser beams, ensure the detection accuracy of the array magnetometer. It is important to carry out related studies of laser beam splitter technology for magnetometer development.

A Deformable Split Fusion Method for Object Detection in High-Resolution Optical Remote Sensing Image

To better address the challenges of complex backgrounds, varying object sizes, and arbitrary orientations in remote sensing object detection tasks, this paper proposes a deformable split fusion method based on an improved RoI Transformer called RoI Transformer-DSF. Specifically, the deformable split fusion method contains a deformable split module (DSM) and a space fusion module (SFM). Firstly, the DSM aims to assign different receptive fields according to the size of the remote sensing object and focus the feature attention on the remote sensing object to capture richer semantic and contextual information. Secondly, the SFM can highlight the spatial location of the remote sensing object and fuse spatial information of different scales to improve the detection ability of the algorithm for objects of different sizes. In addition, this paper presents the ResNext_Feature Calculation_block (ResNext_FC_block) to build the backbone of the algorithm and modifies the original regression loss to the KFIoU to improve the feature extraction capability and regression accuracy of the algorithm. Experiments show that the mAP0.5 of this method on DOTAv1.0 and FAIR1M (plane) datasets is 83.53% and 44.14%, respectively, which is 3% and 1.87% higher than that of the RoI Transformer, and it can be applied to the field of remote sensing object detection.

Swin-FER: Swin Transformer for Facial Expression Recognition

The ability of transformers to capture global context information is highly beneficial for recognizing subtle differences in facial expressions. However, compared to convolutional neural networks, transformers require the computation of dependencies between each element and all other elements, leading to high computational complexity. Additionally, the large number of model parameters need extensive data for training so as to avoid overfitting. In this paper, according to the characteristics of facial expression recognition tasks, we made targeted improvements to the Swin transformer network. The proposed Swin-Fer network adopts the fusion strategy from the middle layer to deeper layers and employs a method of data dimension conversion to make the network perceive more spatial dimension information. Furthermore, we also integrated a mean module, a split module, and a group convolution strategy to effectively control the number of parameters. On the Fer2013 dataset, an in-the-wild dataset, Swin-Fer achieved an accuracy of 71.11%. On the CK+ dataset, an in-the-lab dataset, the accuracy reached 100%.

Beyond singular prototype: A prototype splitting strategy for few-shot medical image segmentation

In the realm of medical image semantic segmentation, few-shot learning, characterized by its efficient data utilization and flexible generalization capabilities, has been garnering increasing attention. The mainstream methods currently employ prototype-based approaches, which extract semantic knowledge from the annotated support images to guide the segmentation of the query image via masked global average pooling. However, such masked global average pooling leads to severe information loss, which is more problematic for medical images with large numbers of highly heterogeneous background categories. In this work, we propose a prototype splitting module (PSM) to effectively address the issue of semantic information loss in few-shot medical image segmentation. Specifically, PSM iteratively splits the support image masks into set of sub-masks containing segmented regions and unsegmented regions in a self-guided manner. This maximally retains the information within the original semantic classes and better extracts the representations of those classes. Additionally, we devise a multi-level cross attention module (MCAM) that transfers the foreground information from the support images to the query images across different levels to facilitate final segmentation prediction. We validate our method on multiple modal and multi-semantic medical image datasets. Results demonstrate that our approach achieves superior performance over existing state-of-the-art methods. The code has been released on https://github.com/fdngh/PSMnet.

Effect of Nitrogen Management Strategies on Yield, Quality and Nitrogen Uptake by Wheat (Triticum aestivum L.)

There is little knowledge among farmers on the rate of nitrogen-fertilizer application. Therefore, there is a need to determine optimal rate of nitrogen-fertilizer application on growth and yield of wheat. Accordingly, an field experiment was conducted during the winter (Rabi) season at Crop Research Centre the Sardar Vallabhbhai Patel University of Agriculture and Technology, Meerut, Uttar Pradesh, to study the impact of different nitrogen level and splitting modules on quality, yield and nitrogen uptake by wheat (Triticum aestivum L.) crop. The main object of investigation was to find out the optimum dose and splitting of nitrogen for wheat in irrigated system under integrated nitrogen management approach. The experiment consisting of four nitrogen levels viz. control i.e. 0 kg N ha-1 , 90 kg N ha-1, i.e 120 kg N ha-1 and 150 kg N ha-1 as main plots and four splitting modules M1- 50% N as basal through VC, + 25% at CRI + 25% booting stage through urea, M2-50 % N as basal through FYM + 25% at tillering + 25% at booting stage through urea, M3 -50% N as basal through VC+FYM (1:1), + 25% at tillering + 20% at booting stage and 5 % N as foliar spray at pre flowering stage through urea, M4 -50% N as basal through VC+FYM (1:1), rest based on SPAD reading through urea as sub-plots. The experimental results revealed that the plot which fertilized with 150 kg N ha-1 gave higher yield-attributing characters and yield (grain, straw and biological yield). The application of 150 kg N ha-1 resulted in 0.44, 15.25 and 80.8 per cent during 2016-17 and 0.41, 13.05 and 79.2 per cent during 2017-18 produced the higher grain yield over 120, 90 and 0 kg N ha-1. The content and uptake of nitrogen by grain, straw and total uptake as well as protein content and protein yield were obtained with 150 Kg N ha-1 but differences between 120 and 150 kg N ha-1 were non-significant, during both the year 2016-17 and 2017-18. Among different splitting modules M3 resulted significantly higher entire yield attributes, yield (grain and straw yield), nitrogen content and uptake by grain, straw and total uptake as well as content and yield of protein. The M3 module produced higher grain yield by 7.9, 14.7 and 24.8 per cent during 2016-17 and 8.1, 14.89 and 25.2 per cent during 2017-18 over M4, M1 and M2 splitting modules, respectively.

A transformer-based pyramid network for coronary calcified plaque segmentation in intravascular optical coherence tomography images

Characterizing coronary calcified plaque (CCP) provides essential insight into diagnosis and treatment of atherosclerosis. Intravascular optical coherence tomography (OCT) offers significant advantages for detecting CCP and even automated segmentation with recent advances in deep learning techniques. Most of current methods have achieved promising results by adopting existing convolution neural networks (CNNs) in computer vision domain. However, their performance can be detrimentally affected by unseen plaque patterns and artifacts due to inherent limitation of CNNs in contextual reasoning. To overcome this obstacle, we proposed a Transformer-based pyramid network called AFS-TPNet for robust, end-to-end segmentation of CCP from OCT images. Its encoder is built upon CSWin Transformer architecture, allowing for better perceptual understanding of calcified arteries at a higher semantic level. Specifically, an augmented feature split (AFS) module and residual convolutional position encoding (RCPE) mechanism are designed to effectively enhance the capability of Transformer in capturing both fine-grained features and global contexts. Extensive experiments showed that AFS-TPNet trained using Lovasz Loss achieved superior performance in segmentation CCP under various contexts, surpassing prior state-of-the-art CNN and Transformer architectures by more than 6.58% intersection over union (IoU) score. The application of this promising method to extract CCP features is expected to enhance clinical intervention and translational research using OCT.

Hierarchical transformer speech depression detection model research based on Dynamic window and Attention merge.

Depression Detection of Speech is widely applied due to its ease of acquisition and imbuing with emotion. However, there exist challenges in effectively segmenting and integrating depressed speech segments. Multiple merges can also lead to blurred original information. These problems diminish the effectiveness of existing models. This article proposes a Hierarchical Transformer model for speech depression detection based on dynamic window and attention merge, abbreviated as DWAM-Former. DWAM-Former utilizes a Learnable Speech Split module (LSSM) to effectively separate the phonemes and words within an entire speech segment. Moreover, the Adaptive Attention Merge module (AAM) is introduced to generate representative feature representations for each phoneme and word in the sentence. DWAM-Former also associates the original feature information with the merged features through a Variable-Length Residual module (VL-RM), reducing feature loss caused by multiple mergers. DWAM-Former has achieved highly competitive results in the depression detection dataset DAIC-WOZ. An MF1 score of 0.788 is received in the experiment, representing a 7.5% improvement over previous research.

Research on a Data-Driven Modeling Method for Precast Concrete Balcony Components

In this paper, a data-driven modeling method for precast concrete (PC) balcony components was proposed to solve the problems of low informatization and the difficult modeling of components at the design stage. Through the analysis of the characteristics of PC balcony components and the combination of modular design methods, the paper designed a data structure for the components and developed a data-driven modeling tool for PC balcony components that can realize the input of structural design data, automatically generating component models. First, this paper introduced the data-driven modeling concept and the modeling process. Second, the PC balcony components in common prefabricated residential projects were analyzed to identify their characteristics. By using a modular design approach, these components were divided and a module dataset was created based on the split modules. Consequently, a data structure for the prefabricated balcony component model was established, wherein both conventional parameters and adaptive parameters between modules were interrelated. Finally, the function of data-driven modeling was achieved by developing a modular design tool on the Revit platform using the C# programming language. The application conducted on a prefabricated building project demonstrated that the software tool and modeling method in this paper effectively improve the level of informatization and modeling efficiency of PC balcony components. The modular design approach was satisfied with the standardization and diversification requirements of balcony components, thereby offering insights for modeling other complex components.

Weakly supervised multi-class semantic video segmentation for road scenes

Weakly supervised multi-class video segmentation is one of the most challenging yet least studied research problems in computer vision. This study aims to investigate two main items: (1) effective feature update for temporal changes combined with feature reuse between temporal frames; and (2) learn object patterns in complex scenes specifically for videos under weak supervision. Associating image tags to visual appearance is not a straightforward learning task, especially for complex scenes. Therefore, in this paper, we present manifold augmentations to obtain reliable pixel labels from image tags. We propose a framework comprised of two key modules: a temporal split module for efficient video processing and a pseudo per-pixel seed generation module for precise pixel-level supervision. Particularly, in our model, we utilize and explore temporal correlations via temporal split module and temporal attention. To reuse the extracted features and incorporate temporal updates for precise and fast computation, a channel-wise temporal split mechanism between successive video frames is presented. Furthermore, we evaluated proposed modules in two additional settings: (1) fully or sparsely supervised road scene video segmentation; and (2) weakly supervised segmentation for complex road scene images. Experiments are conducted on the Cityscapes and CamVid datasets, using DeepLabv3 as segmentation network and LiteFlowNet to compute motion vectors.

HMFT: Hyperspectral and Multispectral Image Fusion Super-Resolution Method Based on Efficient Transformer and Spatial-Spectral Attention Mechanism.

Due to the imaging mechanism of hyperspectral images, the spatial resolution of the resulting images is low. An effective method to solve this problem is to fuse the low-resolution hyperspectral image (LR-HSI) with the high-resolution multispectral image (HR-MSI) to generate the high-resolution hyperspectral image (HR-HSI). Currently, the state-of-the-art fusion approach is based on convolutional neural networks (CNN), and few have attempted to use Transformer, which shows impressive performance on advanced vision tasks. In this paper, a simple and efficient hybrid architecture network based on Transformer is proposed to solve the hyperspectral image fusion super-resolution problem. We use the clever combination of convolution and Transformer as the backbone network to fully extract spatial-spectral information by taking advantage of the local and global concerns of both. In order to pay more attention to the information features such as high-frequency information conducive to HR-HSI reconstruction and explore the correlation between spectra, the convolutional attention mechanism is used to further refine the extracted features in spatial and spectral dimensions, respectively. In addition, considering that the resolution of HSI is usually large, we use the feature split module (FSM) to replace the self-attention computation method of the native Transformer to reduce the computational complexity and storage scale of the model and greatly improve the efficiency of model training. Many experiments show that the proposed network architecture achieves the best qualitative and quantitative performance compared with the latest HSI super-resolution methods.

A novel clean hydrogen production system combining cascading solar spectral radiation and copper-chlorine cycle: Modeling and analysis

In this work, a novel integrated hydrogen production system was proposed. A detailed analysis of the efficiency limit, energy requirement, and parametric impact was theoretically performed. The system combined cascading solar spectral radiation with a copper-chlorine thermochemical cycle, which effectively converted different grades of solar energy to clean hydrogen. The process of a four-step copper-chlorine cycle was simulated to obtain the energy demands using Aspen Plus software. A model for cascading utilization of solar spectral radiation was developed to select the optimal segmented wavelengths and investigate the effects of parameters on the system efficiency. The system designs integrated with other solar cells were also optimized using this model. Among the parameters, the segmented wavelengths were the key parameters that determined the amount of solar energy converted to heat and electricity. The theoretical calculations and simulations were used to determine the performance of this new hydrogen production system. Efficiency analysis demonstrated that the proposed system achieved an ideal efficiency of 68.20%, exceeding the theoretical limit of the single-junction photovoltaic-electrolysis water system. The efficiency of the actual design was 43.44%, higher than the systems without a spectrum splitting module. This study provides an efficient method for realizing high efficiency and high-grade solar energy cascading utilization for producing clean hydrogen energy.

Dense channel splitting network for MR image super-resolution

High spatial resolution is desirable in magnetic resonance imaging (MRI) as it can provide detailed anatomical information, facilitating radiologists with accurate quantitative analysis. Super-resolution (SR) algorithms are effective approaches to enhance MR images' spatial resolution. In the past few years, convolutional neural network (CNN)-based SR methods have significantly improved and outperformed conventional ones. However, existing CNN-based SR methods usually do not explicitly consider the frequency property of images, leading to the limited representation of high-frequency components reflecting image details. To alleviate this problem, a dense channel splitting network (DCSN) algorithm is proposed to process the frequency bands for better feature detection. Specifically, a channel splitting module, a cascaded multi-branch dilation module, and a dense-in and recursive-out mechanism are designed to separate frequency bands of MR images and forward the high-frequency information to deeper layers for reconstruction. Several experiments are performed on real T2 brain and PD (proton density) knee images. The results indicate that the proposed network is superior to conventional CNN-based SR methods.

Module decompositions by images of fully invariant submodules

Let R be a ring with identity, M be a right R-module and F be a fully invariant submodule of M. The concept of an F-inverse split module M has been investigated recently. In this paper, we approach to this concept with a different perspective, that is, we deal with a notion of an F-image split module M, and study various properties and obtain some characterizations of this kind of modules. By means of F-image split modules M, we focus on modules M in which fully invariant submodules F are dual Rickart direct summands. In this way, we contribute to the notion of a T-dual Rickart module M by considering Z?2 (M) as the fully invariant submodule F of M. We also deal with a notion of relatively image splitness to investigate direct sums of image split modules. Some applications of image split modules to rings are given.

A Highly Integrated Automatic Fiber Optical Gyroscope Sensing Coil Winding System

Fiber optic gyroscope (FOG) is a new type of optical sensor used to measure the rotating angular velocity. As the core component, fiber optic coil (FOC) plays a decisive role in the FOG output precision. The manufacturing level and efficiency of FOC have been main factors restricting the application and development of high precision FOG. Precision mechanical positioning method based on the photoelectric devices was researched, and high-performance fully automatic micro mechanical error compensation technique was developed. On these basis. automatic fiber distribution and winding control technology were realized. Eventually a multifunctional system was developed with fiber splitting module, fiber winding module, and calibration module. This system could accomplish automatic preparation of FOC with quadrupolar symmetric pattern according to the parameters set by user. The principle and working flow of each functional module are discussed in details. Experiments show that the system can achieve automatic speed changing, automatic reversing, constant small tension control and precise fiber arrangement. The new designed system greatly improve FOC production efficiency and is conductive to promoting high level FOG development.

Mode-Based Analysis and Optimal Operation of MSF Desalination System

Multi-stage flash (MSF) desalination plays an important role in achieving large-scale fresh water driven by thermal energy. In this paper, based on first-principle modeling of a typical multi-stage flash desalination system, the effects of different operational parameters on system performance and operational optimization for cost saving were extensively studied. Firstly, the modelled desalination system was divided into flash chamber modules, brine heater modules, mixed modules and split modules, and based on energy and mass conservation laws the equations were formulated and put together to describe the whole process model. Then, with physical parameter calculation the whole process was simulated and analyzed on the platform of MATLAB, and the water production performance effected by operational parameters such as the feed temperature of seawater, the recycle brine from the discharge section, steam temperature and flowrate of recycled brine were discussed and analyzed. Then, the optimal operation to achieve maximize GOR (gained output ratio) with fixed freshwater demand was considered and performed, and thus the optimal flowrate of recycled brine, steam temperature, and seawater output flowrate from rejection section were obtained based on the established model. Finally, considering that minimizing the daily operational cost is a more rational objective, the operational cost equations were formulated and the optimal problem to minimize the daily operational cost was solved and the optimal manipulated variables at different hours were obtained. The study results can be used for guideline of real time optimization of the MSF system.

Controlled engineering of nickel carbide induced N-enriched carbon nanotubes for hydrogen and oxygen evolution reactions in wide pH range

The design of cost-effective and competent electrochemical water splitting module that is functional in a wide pH range is still a formidable challenge for viable H2 production. Herein, we report control synthesis of Ni/Ni3C nanoparticles (NPs) induced engineering of the nitrogen-enriched carbon-nanotubes (NCNT) of three different sizes/morphologies by the pyrolysis of melamine with Ni(NO3)·6H2O at various stoichiometric ratios (0.5-wt%, 2-wt% and 4-wt% of Ni). The catalytic potential and structural integrity of the in-situ self-assembled hybrid material was explored and found very effective to catalyse both OER and HER in a broad pH range (0-14). Promisingly, the nanostructuring of the Ni/Ni3C NPs and subsequently the diameter of the NCNT hold the catalytic potential, where 0.5-Ni/Ni3C@NCNT (average size ≈ 15–20 nm) showed remarkable activity for OER (ƞ10-base/acid/neutral = 280/400/600 mV, Tafel slopebase = 42 mV/dec) and HER (ƞ10-acid/base/neutral = 210/290/370 mV, Tafel slopeacid = 40 mV/dec). It is anticipated, that favourable electronic modulation of Ni/Ni3C, structural strain of NCNT, and possible concerted synergistic effect of the hybrid/doped nanocarbons with metal-carbide hold a great potential in this regard, and such modular strategies tailoring the structural-performance relationship may provide the guidance towards the designing of commercially viable electrocatalysts in this regard.

A Lightweight Network Based on Pyramid Residual Module for Human Pose Estimation

The human pose estimation is one of the most popular research fields. Its current accuracy is satisfactory in some cases, however, there exists a challenge for practical application due to the limited memory and computational efficiency in FPGAs and other hardware. We propose a lightweight module based on the pyramid residual module in this work. We change the convolution mode by using the depth-wise separable convolutions structure. Meanwhile, the channel split module and channel shuffle module are added to change the feature graph dimension. As a result, the parameters of the network are reduced effectively. We test the network on standard benchmarks MPII dataset, our method reduces about 50% of the training storage space while maintaining comparable accuracy. The complexity is simplified from 9 GFLOPs to 3 GFLOPs.

A Split-Central-Buffered Load-Balancing Clos-Network Switch With In-Order Forwarding

We propose a configuration scheme for a load-balancing Clos-network packet switch that has split central modules and buffers in between the split modules. Our split-central-buffered Load-Balancing Clos-network (LBC) switch is cell based. The switch has four stages, namely input, central-input, central-output, and output stages. The proposed configuration scheme uses a pre-determined and periodic interconnection pattern in the input and split central modules to load-balance and route traffic. The LBC switch has low configuration complexity. The operation of the switch includes a mechanism applied at input and split-central modules to forward cells in sequence. The switch achieves 100\% throughput under uniform and nonuniform admissible traffic with independent and identical distributions (i.i.d.). These high switching performance and low complexity are achieved while performing in-sequence forwarding and without resorting to memory speedup or central-stage expansion. Our discussion includes throughput analysis, where we describe the operations that the configuration mechanism performs on the traffic traversing the switch, and proof of in-sequence forwarding. A simulation study is presented as a practical demonstration of the switch performance on uniform and nonuniform i.i.d. traffic.

A dual approach to the theory of inverse split modules

Let [Formula: see text] be an arbitrary ring with identity, [Formula: see text] a right [Formula: see text]-module and [Formula: see text] a fully invariant submodule of [Formula: see text]. The notion of an [Formula: see text]-inverse split module [Formula: see text] has been defined and studied by the present authors recently. In this paper, we introduce its dual notion, namely, dual [Formula: see text]-inverse split module [Formula: see text]. This work is devoted to investigation of various properties and characterizations of a dual [Formula: see text]-inverse split module [Formula: see text]. We include applications for rings and cosingular submodules. We also deal with the notion of relatively dual inverse splitness to investigate direct sums of dual inverse split modules.