Ferrimagnetic materials exhibiting remanence can be used to achieve unidirectional electromagnetic-field propagation in the form of magnetoplasmons (MPs) in the subwavelength regime. This study investigates the MP properties and various guiding modes in a hollow cylindrical waveguide made of materials that exhibit remanence. Pattern analysis and numerical simulations are used to demonstrate that dispersion relationships and electromagnetic-field distribution are strongly affected by the operating frequency and physical dimensions of the structure. In addition, the existence of two different guiding modes is proved, namely regular and surface-wave modes. By adjusting the operating frequency and reducing the diameter of the hollow cylinder, the regular mode can be suppressed so as to only retain the surface-wave mode, which enables unidirectional MP propagation in the cylindrical waveguide. Moreover, the unidirectional surface-wave mode is robust to backscattering due to surface roughness and defects, which makes it very useful for application in field-enhancement devices.

In this paper, we have mainly studied the amplification effect of thulium-doped fiber amplifier (TDFA) at 2 µm, and compared different amplification effects of the one-stage TDFA, two-stage TDFA and three-stage TDFA at proper conditions. The simulation results show that within the effective threshold, with the increase of the pump power, the amplification effect of the optical amplifier improves, but the signal-to-noise ratio (SNR) of the output signal decreases, in order to balance the gain benefit and noise coefficient of TDFA, we can use a multi-stage amplification structure. Three-stage backward-pumped series 2.06 µm TDFA, whose slope efficiency can achieve 11% at certain condition. At 5.2 W pump power, the output signal gain of 2 µm TDFA exceeds 20 dB, and the output SNR is higher than 32 dB. In addition, the effect of the optimum length of thulium-doped fiber on the amplification performance of 2 µm TDFA is also analyzed in this paper. These simulation results are important for the experiment and design of 2 µm TDFA.

Based on optofluidics and whispering gallery mode (WGM) theory, here an optofluidic refractive index sensor with microtube-coupled suspended core fiber (SCF) is proposed. It solves the issues of general sensors with microcavity-coupled fiber taper such as too fragile, unstable performance due to open coupling, poor portability and repeatability, while overcoming the poor performance of low refractive index sensing in general full-package fiber sensors. The sensor only needs a very small amount of liquid sample (about 1.8 nL). The proposed sensor combines the excellent performance of full package, optofluidics and WGM resonator. The resonant characteristics and sensing performance of the sensor are analyzed and discussed by the theoretical simulation. The simulation results indicate that the sensor has a wide refractive index sensing range (1.330–1.700) and good performance. The resonance wavelength shift has a good linear relationship with the liquid refractive index variation. In the low refractive index region, the sensitivity is 222.5–247.5 nm/RIU, Q-factor is 1.03×10³ and the detection limit is 3.64×10⁻⁴ RIU. In the medium and high refractive index regions, the sensitivity is 564.4–846.2 nm/RIU, Q-factor is up to 8.62×10⁴, and the detection limit can be as low as 1.29×10⁻⁶ RIU. The sensor exhibits a high sensitivity, a high Q-factor and a very low detection limit.

In this paper, we present a novel approach to enhancing the visible light photodetection efficiency of reduced graphene oxide (rGO) by incorporating polypyrrole (Ppy) nanoparticles sized between 126 nm and 1 025 nm. The rGO and Ppy nanoparticles were synthesized via Hummer’s method and chemical polymerization, respectively. Characterization was performed using scanning electron microscope (SEM), transmission electron microscope (TEM), Raman spectroscopy, and optical measurements. The rGO/Ppy photodetector demonstrated a high photoresponsivity of 15 mA/W and a broad spectral response from 405 nm to 805 nm, indicating improved efficiency and versatility. This study high-lights the potential of tailored Ppy nanoparticle sizes in advancing rGO photodetectors for high-performance optoelectronic applications.

Organosilicone thin films were prepared through plasma polymerization (pp) in a plasma enhance chemical vapour deposition (PECVD) system, utilizing hexamethyldisilazane (HMDSN) as a monomer precursor, at varying distances (25 mm, 35 mm, 45 mm, 55 mm, and 65 mm) from the plasma source to the substrate. Research has examined how the distance between the substrate and plasma source impacts the properties of thin films, including their thickness, surface morphology, and photoluminescence (PL). It was discovered that as the distance increased, both film thickness and PL intensity also increased. Additionally, the film was observed to be more uniform and smoother when deposited 45 mm below the plasma source.

Power transmission lines are a critical component of the entire power system, and ice accretion incidents caused by various types of power systems can result in immeasurable harm. Currently, network models used for ice detection on power transmission lines require a substantial amount of sample data to support their training, and their drawback is that detection accuracy is significantly affected by the inaccurate annotation among training dataset. Therefore, we propose a transformer-based detection model, structured into two stages to collectively address the impact of inaccurate datasets on model training. In the first stage, a spatial similarity enhancement (SSE) module is designed to leverage spatial information to enhance the construction of the detection framework, thereby improving the accuracy of the detector. In the second stage, a target similarity enhancement (TSE) module is introduced to enhance object-related features, reducing the impact of inaccurate data on model training, thereby expanding global correlation. Additionally, by incorporating a multi-head adaptive attention window (MAAW), spatial information is combined with category information to achieve information interaction. Simultaneously, a quasi-wavelet structure, compatible with deep learning, is employed to highlight subtle features at different scales. Experimental results indicate that the proposed model in this paper outperforms existing mainstream detection models, demonstrating superior performance and stability.

This paper proposes a novel method for the automatic diagnosis of keratitis using feature vector quantization and self-attention mechanisms (ADK_FVQSAM). First, high-level features are extracted using the DenseNet121 backbone network, followed by adaptive average pooling to scale the features to a fixed length. Subsequently, product quantization with residuals (PQR) is applied to convert continuous feature vectors into discrete features representations, preserving essential information insensitive to image quality variations. The quantized and original features are concatenated and fed into a self-attention mechanism to capture keratitis-related features. Finally, these enhanced features are classified through a fully connected layer. Experiments on clinical low-quality (LQ) images show that ADK_FVQSAM achieves accuracies of 87.7%, 81.9%, and 89.3% for keratitis, other corneal abnormalities, and normal corneas, respectively. Compared to DenseNet121, Swin transformer, and InceptionResNet, ADK_FVQSAM improves average accuracy by 3.1%, 11.3%, and 15.3%, respectively. These results demonstrate that ADK_FVQSAM significantly enhances the recognition performance of keratitis based on LQ slit-lamp images, offering a practical approach for clinical application.

Accessible communication based on sign language recognition (SLR) is the key to emergency medical assistance for the hearing-impaired community. Balancing the capture of both local and global information in SLR for emergency medicine poses a significant challenge. To address this, we propose a novel approach based on the inter-learning of visual features between global and local information. Specifically, our method enhances the perception capabilities of the visual feature extractor by strategically leveraging the strengths of convolutional neural network (CNN), which are adept at capturing local features, and visual transformers which perform well at perceiving global features. Furthermore, to mitigate the issue of overfitting caused by the limited availability of sign language data for emergency medical applications, we introduce an enhanced short temporal module for data augmentation through additional subsequences. Experimental results on three publicly available sign language datasets demonstrate the efficacy of the proposed approach.