Rolling bearing fault diagnosis method based on gramian angular difference field and dynamic self-calibrated convolution module.

To address the problem of insufficient feature extraction abilities of traditional fault diagnosis methods under conditions of sample scarcity and strong noise interference, a rolling bearing fault diagnosis method based on the Gramian Angular Difference Field (GADF) and Dynamic Self-Calibrated Convolution (DSC) is proposed. First, the GADF method converts one-dimensional signals into GADF images to capture nonlinear relationships and periodic information in time-series data. Second, a dynamic self-calibrated convolution module is introduced to enhance the feature extraction ability of the model.

Feature optimization-guided high-precision and real-time metal surface defect detection network.

Existing computer vision-based surface defect detection techniques for metal materials typically encounter issues with defect overlap, significant differences within classes, and similarity between defect samples. These issues compromise feature extraction accuracy and result in missed and false detections. This study proposed a feature optimization-guided high-precision and real-time metal surface defect detection network (FOHR Net) to improve defect feature expressiveness.

Multiple Magnetic Phase Transitions and Giant Refrigerant Capacity in a GdDyHoErTm High-Entropy Alloy.

Magnetocaloric high-entropy alloys (HEAs) have recently garnered significant interest owing to their potential applications in magnetic refrigeration, offering a wide working temperature range and large refrigerant capacity. In this study, we thoroughly investigated the structural, magnetic, and magnetocaloric properties of equiatomic GdDyHoErTm HEAs. The as-cast alloy exhibits a single hexagonal phase, a randomly distributed grain orientation, and complex magnetism.

A non-local dual-stream fusion network for laryngoscope recognition.

Purpose: To use deep learning technology to design and implement a model that can automatically classify laryngoscope images and assist doctors in diagnosing laryngeal diseases.

Materials And Methods: The experiment was based on 3057 images (normal, glottic cancer, granuloma, Reinke's Edema, vocal cord cyst, leukoplakia, nodules and polyps) from the dataset Laryngoscope8. A classification model based on deep neural networks was developed and tested.

Sunlight-Drivable Composite Film Using Carbon Nanopowder-doped PVDF and Liquid Crystal Polymer Network.

Actuators based on liquid crystals have garnered significant attention due to their potential applications in wearable technology and bionic soft robots. Composite films composed of liquid crystal polymer networks (LCNs) and other stimulus-responsive materials exhibit the capability to convert external stimuli into mechanical deformation. However, the development of sunlight-driven actuators presents significant challenges, primarily due to the relatively low intensity of sunlight and the limited conversion efficiency of photothermal materials.

Direct Implementation of MS Using Frequency Scanning Collision Induced Dissociation in a Digital Ion Trap Mass Spectrometer.

Tandem mass spectrometry (MS) is one of the most effective methods to obtain the structures of organic molecules, enabling the observation of multigenerational ion fragments. Collision-induced dissociation (CID) is currently the most mature technique for mass spectrometry analysis. Ion trap mass spectrometry (ITMS) is favored for on-site detection field, due to its ability of MS analysis with a single trap and its small size.

MSA-MaxNet: Multi-Scale Attention Enhanced Multi-Axis Vision Transformer Network for Medical Image Segmentation.

Convolutional neural networks (CNNs) are well established in handling local features in visual tasks; yet, they falter in managing complex spatial relationships and long-range dependencies that are crucial for medical image segmentation, particularly in identifying pathological changes. While vision transformer (ViT) excels in addressing long-range dependencies, their ability to leverage local features remains inadequate. Recent ViT variants have merged CNNs to improve feature representation and segmentation outcomes, yet challenges with limited receptive fields and precise feature representation persist.

A literature review of artificial intelligence (AI) for medical image segmentation: from AI and explainable AI to trustworthy AI.

Background And Objective: Medical image segmentation is a vital aspect of medical image processing, allowing healthcare professionals to conduct precise and comprehensive lesion analyses. Traditional segmentation methods are often labor intensive and influenced by the subjectivity of individual physicians. The advent of artificial intelligence (AI) has transformed this field by reducing the workload of physicians, and improving the accuracy and efficiency of disease diagnosis.

Photothermally-enhanced ferroptotic-chemo therapy enabled by ZIF-derived multizyme.

A multi-functional single-Fe-atom nanozyme (Fe-SAzyme) is designed, integrating the near-infrared photothermal property, the ability to carry chemoagent (doxorubicin - DOX), and nanocatalytic activities mimicking peroxidase, oxidase, and glutathione oxidase. The nanocatalytic activities act cooperatively to effectively produce cytotoxic radicals in the tumor microenvironment (TME), thereby leading to ferroptosis of cancer cells. The photothermal effect not only enhances the nanocatalytic therapy but also enables photothermal therapy.

Effect of 2-Mercapto-1-methylimidazole on the Dual Action of Chemical Mechanical Polishing of Cu and Ta.

When chemical mechanical polishing (CMP) of tantalum (Ta)-based barrier layers is done through silicon via (TSV), it is necessary to control the rate selection of copper (Cu) and Ta to prevent the formation of dishing pit formation due to the rapid removal rate (RR) of copper in the via compared to tantalum outside the via. This paper selected 2-mercapto-1-methylimidazole (TAMZ) as an inhibitor, which has the dual effect of reducing the RR of Cu and increasing the RR of Ta. The experimental results show that the rate selection ratio of Cu and Ta is up to 2.

Spintronic terahertz metasurface emission characterized by scanning near-field nanoscopy.

Understanding the ultrafast excitation, detection, transportation, and manipulation of nanoscale spin dynamics in the terahertz (THz) frequency range is critical to developing spintronic THz optoelectronic nanodevices. However, the diffraction limitation of the sub-millimeter waves - THz wavelengths - has impaired experimental investigation of spintronic THz nano-emission. Here, we present an approach to studying laser THz emission nanoscopy from W|CoFeB|Pt metasurfaces with ∼60-nm lateral spatial resolution.

Ultra-High Capacitive Energy Storage Density at 150 °C Achieved in Polyetherimide Composite Films by Filler and Structure Design.

Polymer dielectrics are crucial for electronic communications and industrial applications due to their high breakdown field strength (E), fast charge/discharge speed, and temperature stability. The upcoming electronic-electrical systems pose a significant challenge, necessitating polymeric dielectrics to exhibit exceptional thermal stability and energy storage capabilities at high temperatures. Here, ultra-high dielectric constant (ɛ) and charge/discharge efficiency (η) of 0.

A rapid, non-destructive, and accurate method for identifying citrus granulation using Raman spectroscopy and machine learning.

Citrus fruits are widely consumed for their nutritional value and taste; however, juice sac granulation during fruit storage poses a significant challenge to the citrus industry. This study used Raman spectroscopy coupled with machine learning algorithms to rapidly, non-destructively, and precisely detect citrus granulation. The investigation analyzed 969 Raman spectral data points, comprising 714 non-granulated and 255 granulated citrus samples.

A waveguide branch-line crossover based on metal gratings.

A waveguide branch-line crossover with broad bandwidth is designed for beam-forming networks in millimeter wavebands. To realize the power coupling between two parallel branch lines, a substrate with metal gratings patched replaces the common E-plane waveguide wall. Compared with the traditional crossover with coupling vias, the metal gratings achieve continuous and efficient power coupling, endowing the proposed crossover with advantages in compactness and broad bandwidth.

Ultrasonic wind vector measurement based on quadratic correlation two-frequency phase method.

A proposed method for measuring ultrasonic wind speed and direction utilizes the quadratic correlation phase method to address limited range issues and low measurement accuracy associated with the phase difference method when the signal-to-noise ratio is low. First, the proposed method uses quadratic correlation method combined with the four-step phase shift method to estimate the phase difference based on wind measuring structure of the opposing-type Array Structures. Then, the two-frequency phase difference method extends the measurement range of ultrasonic wind speed and direction.

Novel terahertz optical switch based on PIT phenomenon and Lorentz theory.

We propose and demonstrate a structure consisting of graphene rings and square rings that enables broadband and tunable plasmon-induced transparency (PIT) effects. Through coupled Lorentz model analysis, we attribute the transmission window at 2.1 THz to the interference between the equipartitioned exciton resonance of the graphene ring pairs and the inductive-capacitive resonance of the graphene square ring pairs.

Study on the travelers' P&R usage behavior during snowfall in the corridor valley pattern city.

According to the characteristics of corridor valley pattern cities, this paper presents four aspects of terrain space perception. Using a structural equation model (SEM), the four-dimensional terrain space perception is analyzed through second-order confirmatory factor analysis. A random parameter Logit model (RPLM) was developed to examine the impact of individual socioeconomic attributes, objective built environment, travel characteristics, and terrain space perception on travelers' Park and Ride (P&R) use behavior during snowfall conditions.

Spike-VisNet: A novel framework for visual recognition with FocusLayer-STDP learning.

Current vision-inspired spiking neural networks (SNNs) face key challenges due to their model structures typically focusing on single mechanisms and neglecting the integration of multiple biological features. These limitations, coupled with limited synaptic plasticity, hinder their ability to implement biologically realistic visual processing. To address these issues, we propose Spike-VisNet, a novel retina-inspired framework designed to enhance visual recognition capabilities.

Cholesteric Liquid Crystal Network Composite Film with Structure, Dual-Band Responsive Luminescent Color for Multiple Anti-Counterfeiting and Information Encryption.

Anti-counterfeiting technology plays an indispensable role in the high-tech field and various critical application areas. However, traditional anti-counterfeiting approaches currently in widespread use are too simplistic and easily replicated or forged, while advanced technologies with multiple anti-counterfeiting functions remain in the developmental stage. This study presents a novel multiple anti-counterfeiting technique.

Pure spin current generation with photogalvanic effects in h-BN/graphene/h-BN van der Waals vertical heterostructures.

We have computationally demonstrated a new method for generating pure spin current with the photogalvanic effect (PGE) by constructing transport junctions using h-BN/graphene/h-BN van der Waals (vdW) heterostructure leads. It has been observed that the pure spin current without any accompanying charge current induced by the PGE can consistently be obtained, regardless of photon energy and polarization/helicity angle, as well as the specific type of polarization (linear, circular, or elliptical). The mechanism lies in the structural inversion symmetry and real space spin polarization antisymmetry of the junctions.

ASIP Performance Enhancement by Hazard Control through Scoreboard.

The application-specific instruction set processor (ASIP) has been gradually accepted in AI, communication, media, game and industry control. The digital signal processor (DSP) is a typical ASIP, whose benefits include high performance in specific domains, low power consumption, high flexibility and low silicon consumption. One of the challenges for DSP design is to handle problems induced by datapath acceleration.