The search for graphene or few layer graphene production methods that are simple, allow mass production and yield good quality material continues to provoke intense investigation. The present work contributes through the study of the aqueous exfoliation of four types of graphene sources, namely graphite and graphite nanoflakes with different morphologies and geographical origin. The exfoliation was achieved in an aqueous solution of a soluble pyrene derivative that was synthesized to achieve maximum interaction with the graphene surface at low concentration (5 x 10-5 M). The yield of bilayer and few layer graphene obtained was quantified by Raman spectroscopic analysis and the adsorption of the pyrene derivative on the graphene surface was studied by thermogravimetric analysis and X-ray diffraction. The whole procedure was rationalized with the help of molecular modeling.

Graphene nanostructures (GNS) are among the most promising materials for creating supercapacitors. However, GNS are still not used in creating supercapacitors due to the impossibility of obtaining large volumes of high-quality material at an acceptable cost. In this work, we investigated the efficiency of using few-layer graphene (FLG) with the number of layers is not more than 5 synthesized under Self-propagating high-temperature synthesis (SHS) conditions as basic material for supercapacitors. Using the SHS process makes it possible to synthesize large volumes of FLG without Stone-Wales defects. It was found that the synthesized FLG makes it possible to obtain better results than using classical materials, namely activated carbon (AC). It was found that the sample based on FLG has a higher specific capacitance – 65 F×g-1 compared to the sample from AC, the specific capacitance of which is 35 F×g-1, for a speed of 5 mV×s-1 these values are 170 and 64 F×g-1, respectively. The drop in capacitance over 1000 cycles was 4%, which indicates a sufficiently high stability of FLG and allows us to consider FLG as perspective material for use in supercapacitors.

A quantitative method is proposed to determine of Stone-Wales defects for carbon nanostructures with sp2 hybridization of carbon atoms. The technique is based on the diene synthesis reaction (Diels-Alder reaction). The proposed method was used to determine Stone-Wales defects in the few-layer graphene (FLG) nanostructures synthesized by the self-propagating high-temperature synthesis (SHS) process, in reduced graphene oxide (rGO) synthesized based on the method of Hammers and in the single-walled carbon nanotubes (SWCNT) TUBAL trademark, Russia. Our research has shown that the structure of FLG is free of Stone-Wales defects, while the surface concentration of Stone-Wales defects in TUBAL carbon nanotubes is 1.1×10-5 mol/m2 and 3.6×10-5 mol/m2 for rGO.

For the first time, few-layer graphene (FLG) nanosheets were synthesized by the method of self-propagating high-temperature synthesis (SHS) from biopolymers (starch and lignin). We suggested that biopolymers (lignin, tree bark) and polysaccharides, in particular starch, could be an acceptable source of native cycles for the SHS process. The carbonization of biopolymers under the conditions of the SHS process was chosen as the basic method of synthesis. Chemical reactions, under the conditions of the SHS process, proceed according to a specific mechanism of nonsothermal branched-chain processes, which are characterized by the joint action of two fundamentally different process-accelerating factors - avalanche reproduction of active intermediate particles and self-heating. The method of obtaining FLG nanosheets included the thermal destruction of hydrocarbons in a mixture with an oxidizing agent. We used biopolymers as hydrocarbons and ammonium nitrate as an oxidizing agent. Thermal destruction was carried out in the mode of SHS, heating the mixture in a vessel at a speed of 20–30 oC/min to 150-200 oC and keeping at this temperature for 15–20 min with the discharge of excess gases into atmosphere. A combination of spectrometric research methods, supplemented by electron microscopy data, has shown that the particles of the carbonated product powder in their morphometric and physical parameters correspond to FLG nanosheets. An X-ray diffraction analysis of the indicated FLG nanosheets was carried out, which showed the absence of formations with a graphite crystal structure in the final material. The surface morphology was also studied and the features of the IR absorption of FLG nanosheets were analyzed. It is shown that the developed SHS method makes it possible to obtain FLG nanosheets with linear dimensions of tens of microns and a thickness of not more than 1-5 graphene layers (several graphene layers).

The results of studying the effect of particles of few-layer graphene (FLG) synthesized by self-propagating high-temperature synthesis (SHS) on the complex of strength and thermo physical properties of polymer composite products obtained by DLP 3D printing are presented. It has been discovered to achieve an increase in thermo physical and strength parameters of polymers modified by FLG compared with samples made on the unmodified base resin. This result was achieved due to low defectiveness, namely the absence of Stone-Wales defects in the structure of FLG due to the homogeneous distribution of FLG over the volume of the polymer in the form of highly dispersed aggregates. It was possible to increase hardness by 120 %, bending strength by 102 %, Charpy impact strength by 205 %, and thermal conductivity at 25 °C by 572 % at concentrations of few-layer graphene of no more than 2 wt. %.

Carbonization of a biopolymer (lignin) in self-propagating high-temperature synthesis (SHS) led to the production of 2D graphene structures. With the help of modern analytical methods (Raman spectroscopy, X-ray diffraction) and electron microscopy, the resulting product was confirmed to have a few-layer 2D graphene structure. The predicted photovoltaic properties of the resulting few-layer graphene were implemented for laser ignition of a model pyrotechnic composition based on porous silicon. A phenomenological model of the formation mechanism of 2D graphene structures under the conditions of the SHS process is proposed.

Chemical process control relies on a tightly controlled, narrow range of margins for critical variables, ensuring process stability and safeguarding equipment from potential accidents. The availability of historical process data is limited to a specific setpoint of operation. This challenge raises issues for process monitoring in predicting and adjusting to deviations outside of the range of operational parameters. Therefore, this paper proposes simulation-assisted deep transfer learning for predicting and optimizing the final purity and production capacity of the glycerin purification process. The proposed network is trained by the simulation domain to generate a base feature extractor, which is then fine-tuned using few-shot learning techniques on the target learner to extend the working domain of the model beyond the historical practice. The result shows that the proposed model improved prediction performance by 99% in predicting water content and 79.72% in glycerin prediction over the conventional deep learning model. Additionally, the implementation of the proposed model identified the production and product quality improvement for enhancing the glycerin purification process.

Few-shot classification is challenging since the goal is to classify unlabeled samples with very few labeled samples provided. It has been shown that cross attention helps generate more discriminative features for few-shot learning. This paper extends the idea and proposes two cross attention modules, namely the cross scaled attention (CSA) and the cross aligned attention (CAA). Specifically, CSA scales different feature maps to make them better matched, and CAA adopts the principal component analysis to further align features from different images. Experiments showed that both CSA and CAA achieve consistent improvements over state-of-the-art methods on four widely used few-shot classification benchmark datasets, miniImageNet, tieredImageNet, CIFAR-FS, and CUB-200-2011, while CSA is slightly faster and CAA achieves higher accuracies.

With the increasing utilization of intelligent mobile devices for online inspection of electrical equipment in smart grids, the limited computing power and storage capacity of these devices pose challenges for deploying large algorithm models and it’s hard to obtain a substantial number of images of electrical equipment in public. In this paper, we propose a novel distillation method that compresses the knowledge of teacher networks into a compact few-shot classification network, employing a global and local knowledge distillation strategy. Central to our method is exploiting the global and local relationship between the features exacted by the backbone of the teacher network and student network. We compare our method with recent state-of-the-art (SOTA) methods on three public datasets and achieve superior performance. Additionally, we contribute a new dataset, namely EEI-100, which is specifically designed for classification of electrical equipment. We validate our method on this dataset and demonstrate its exceptional prediction accuracy of 94.12% when utilizing only 5-shot images.

This work presents results of property researches of fabricated samples of silica few-mode optical fiber (FMF) with induced chirality under twisting 10 and 66 revolutions per meter, core diameter 11 µm, typical “telecommunication” cladding diameter 125 µm and improved height of quasi-step refractive index profile. Proposed FMF supports 4 guided modes over “C”-band. We represent results of computed characteristics, as well as experimentally measured spectral responses of laser-excited optical signals, including researches and analysis of few-mode effects, occurring after fiber Bragg grating writing.

Agriculture is pivotal in national economies, with pest detection significantly influencing food quality and quantity. Pest classification remains challenging in automated agriculture monitoring systems, exacerbated by the non-uniform pest scales and the scarcity of high-quality datasets. In this study, we constructed a pest dataset by acquiring domain-agnostic images from the Internet and resizing them to a standardized 299x299 pixel format. Additionally, we employed diffusion models to generate supplementary data. While Convolutional Neural Networks (CNNs) are prevalent for prediction and classification, they often lack effective global information integration and discriminative feature representation. To address these limitations, we propose the RS Transformer, an innovative model that combines elements like the Region Proposal Network, Swin Transformer, and ROI Align. Additionally, we introduce the Randomly Generated Stable Diffusion Dataset (RGSDD) to augment the availability of high-quality pest datasets. Extensive experimental evaluations demonstrate the superiority of our approach compared to both two-stage models (SSD and Faster R-CNN) and one-stage models (YOLOv3, YOLOv4, YOLOv5m, YOLOv8, and DETR). We rigorously assess performance using metrics such as mean Average Precision (mAP), F1Score, Recall, and mean Detection Time (mDT). Our research contributes to advancing pest detection methodologies in automated agriculture systems, promising improved food production and quality.

Target detection and segmentation in synthetic aperture radar (SAR) images are vital steps for many remote sensing applications. In the era of data-driven deep learning, this task is extremely challenging due to the limited labeled data. Few-shot Learning has the ability to learn quickly from few samples with supervised information. Inspired by this, a few-shot learning framework named MSG-FN is proposed to solve the segmentation of ship targets in heterologous SAR images with few annotated samples. The proposed MSG-FN adopts a dual-branch network consisting of a support branch and a query branch. The support branch is used to extract features with an encoder, and the query branch uses a U-shaped encoder-decoder structure to segment the target in the query image. The encoder of each branch is composed of well-designed residual blocks combined with filter response normalization to capture robust and domain-independent features. A multi-scale similarity guidance module is proposed to improve the scale adaptability of detection by applying hand-on-hand guidance of support features to query features of various scales. In addition, a SAR dataset named SARShip-4i is built to evaluate the proposed MSG-FN and the experimental results show that the proposed method achieves superior segmentation results compared with the state-of-the-arts.

Few-shot object detection (FSOD) aims to address the challenge of requiring a substantial amount of annotations for training in conventional object detection, which is very labor-intensive. However, existing few-shot methods achieve high precision with the sacrifice of time-consuming for exhaustive fine-tuning, or take poor performance in novel-class adaptation. We presume the major reason is that the valuable correlation feature among different categories is insufficiently exploited, hindering the generalization of knowledge from base to novel categories for object detection. In this paper, we propose Few-Shot object detection via Correlation-RPN and Transformer Encoder-Decoder (CRTED), a novel training network to learn object-relevant features of inter-class correlation and intra-class compactness while suppressing object-agnostic features in the background with limited annotated samples. And we also introduce a 4-way tuple-contrast training strategy to positively activate the training progress of our object detector. Experiments over two few-shot benchmarks (Pascal VOC, MS-COCO) demonstrate that, our proposed CRTED without further fine-tuning can achieve comparable performance with current state-of-the-art fine-tuned works. The codes and pre-trained models will be released.

Ransomware attacks have rapidly proliferated, inflicting severe financial damages on businesses and individuals. Machine learning approaches to automate ransomware detection have shown promise but grapple with challenges like limited training data. This study introduces a novel deep learning model for few-shot ransomware classification. The model employs entropy features derived directly from malware binaries coupled with a twin neural network architecture utilizing transfer learning. Tests on over 1000 samples across 11 families demonstrate a weighted F1-score of 85.8\%, surpassing existing methods. The approach mitigates biases in limited training data and preserves intricacies lost in image-based features. It exhibits precise classification capabilities even with sparse samples of new ransomware variants. The research highlights the potential of entropy-driven deep learning to equip defenses against emerging zero-day ransomware strains.

In recent years, different deep learning frameworks were introduced for hyperspectral image (HSI) classification. However, the proposed network models have a higher model complexity and do not provide high classification accuracy if few-shot learning is used. This paper pre-sents an HSI classification method that combines random patches network (RPNet) and re-cursive filtering (RF) to obtain informative deep features. The proposed method first convolves image bands with random patches to extract multi-level deep RPNet features. Thereafter, the RPNet feature set is subjected to dimension reduction through principal component analysis (PCA) and the extracted components are filtered using the RF procedure. Finally, HSI spectral features and the obtained RPNet-RF features are combined to classify the HSI using a support vector machine (SVM) classifier. In order to test the performance of the proposed RPNet-RF method, some experiments were performed on three widely known datasets using a few training samples for each class and classification results were compared with those obtained by other advanced HSI classification methods adopted for small training samples. The comparison showed that the RPNet-RF classification is characterized by higher values of such evaluation metrics as overall accuracy and Kappa coefficient (https://github.com/UchaevD/RPNet-RF).

CNN-based methods have dominated the field of aerial scene classification for the past few years. While achieving remarkable success, CNN-based methods suffer from excessive parameters and notoriously rely on large amounts of training data. In this work, we introduce few-shot learning to the aerial scene classification problem. Few-shot learning aims to learn a model on base-set that can quickly adapt to unseen categories in novel-set, using only a few labeled samples. To this end, we proposed a meta-learning method for few-shot classification of aerial scene images. First, we train a feature extractor on all base categories to learn a representation of inputs. Then in the meta-training stage, the classifier is optimized in the metric space by cosine distance with a learnable scale parameter. At last, in the meta-testing stage, the query sample in the unseen category is predicted by the adapted classifier given a few support samples. We conduct extensive experiments on two challenging datasets: NWPU-RESISC45 and RSD46-WHU. The experimental results show that our method yields state-of-the-art performance. Furthermore, several ablation experiments are conducted to investigate the effects of dataset scale, the impact of different metrics and the number of support shots; the experiment results confirm that our model is specifically effective in few-shot settings.

Few-shot fine-grained image classification (FSFGIC) methods refer to machine learning methods which aim to classify images (e.g., bird species, flowers, and airplanes) belonging to subordinate object categories of the same entry-level category with only a few samples. It is worth to note that feature representation learning is used not only to represent training samples, but also to construct classifiers for performing various FSFGIC tasks. In this paper, starting from the definition of FSFGIC, a taxonomy of feature representation learning for FSFGIC is proposed. According to this taxonomy, we discuss key issues on FSFGIC (including data augmentation, local or/and global deep feature representation learning, class representation learning, and task specific feature representation learning). The existing popular datasets and evaluation standards are introduced. Furthermore, a novel classification performance evaluation mechanism is designed with a 0.95 confidence interval for judging whether the classification accuracy obtained by a certain specified method is good or bad. Moreover, current challenges and future trends of feature representation learning on FSFGIC are elaborated.

Mineral prospectivity mapping (MPM), aiming to outline and prioritize mineral exploration targets, has been spurred by data-driven machine learning algorithms. Supervised data-driven MPM is a typical few-shot task, suffering from the scarcity of labeled data, over-fitting of models and uncertainty of predictions. The main objective of this contribution is to propose a robust framework of few-shot learning (FSL) combining data augmentation and transfer learning, which enables generation of prospectivity models with excellent predictive efficiency and low uncertainty. The mineral systems approach was used to transfer a conceptual mineral system into mappable exploration criteria. Synthetic minority over-sampling technique (SMOTE) was employed to augment and balance the labeled dataset, allowing for model pre-training with a large synthetic training dataset of source domain. The knowledge derived from pre-trained models was then transferred to the target domain by fine-tuning, and the prospectivity model was generated in light of over-fitting and uncertainty assessment. The proposed FSL framework was applied to tungsten prospectivity mapping in southern Jiangxi Province. The results indicate that the SMOTE-ed balanced dataset boosts the classification accuracy in the training process. The FSL models yield an arch-shaped prediction point pattern favorable for focusing potential targets with high probability and low uncertainty. The FSL models achieve a high predictive performance (test AUC=0.9172) and the lowest quantitative over-fitting value, compared to the models derived from the benchmark algorithms of random forest and support vector machine. Four levels of potential targeting zones, considering both predictive efficiency and uncertainty, are extracted from the resulting FSL prospectivity map. The final high-potential and low-risk exploration targets only cover 4.27% of the area, but capture 41.53% known tungsten deposits, achieving superior predictive performance. This study highlights the capability of FSL framework for controlling over-fitting and generating high-confidence exploration targets with low uncertainty.

The rapid growth of smartphone financial services raises the need for secure mobile authentication. Continuous authentication is a user-friendly way to strengthen the security of smartphones by implicitly monitoring a user’s identity through sessions. Mobile continuous authentication can be viewed as an anomaly detection problem in which models discriminate between one genuine user and the rest of the imposters (anomalies). In practice, complete imposter profiles are hardly available due to the time and monetary cost, while leveraging genuine data alone yields poor generalized models due to the lack of knowledge about imposters. To address this challenge, we recast continuous mobile authentication as a few-shot anomaly detection problem, aiming to enhance the inference robustness of unseen imposters by using partial knowledge of available imposter profiles. Specifically, we propose a novel deep learning-based model, namely Local Feature Pooling based Temporal Convolution Network (LFP-TCN), which directly models raw sequential mobile data, aggregating global and local feature information. In addition, we introduce a random pattern mixing augmentation to generate class-unconstrained imposter data for the training. The augmented pool enables characterizing various imposter patterns from limited imposter data. Finally, we demonstrate practical continuous authentication using score-level fusion, which prevents long-term dependency or increased model complexity due to extended re-authentication time. Experiments on two public benchmark datasets show the effectiveness of our method and its state-of-the-art performance.

The central goal of few-shot scene classification is to learn a model that can generalize well to a novel scene category (UNSEEN) from only one or a few labeled examples. Recent works in the remote sensing (RS) community tackle this challenge by developing algorithms in a meta-learning manner. However, most prior approaches have either focused on rapidly optimizing a meta-learner or aimed at finding good similarity metrics while overlooking the embedding power. Here we propose a novel Task-Adaptive Embedding Learning (TAEL) framework that complements the existing methods by giving full play to feature embedding’s dual roles in few-shot scene classification - representing images and constructing classifiers in the embedding space. First, we design a lightweight network that enriches the diversity and expressive capacity of embeddings by dynamically fusing information from multiple kernels. Second, we present a task-adaptive strategy that helps to generate more discriminative representations by transforming the universal embeddings into task-specific embeddings via a self-attention mechanism. We evaluate our model in the standard few-shot learning setting on two challenging datasets: NWPU-RESISC4 and RSD46-WHU. Experimental results demonstrate that, on all tasks, our method achieves state-of-the-art performance by a significant margin.

Leveraging the open-world understanding capacity of large-scale visual-language pre-trained models has become a hot-spot in point cloud classification. Recent approaches rely on transferable visual-language pre-trained models, classifying point clouds by projecting them into 2D images and evaluating consistency with textual prompts. These methods benefit from the robust open-world understanding capabilities of visual-language pre-trained models and require no additional training. However, they face several challenges summarized as prompt ambiguity, image domain gap, view weights confusion, and feature deviation. In response to these challenges, we propose PointBLIP, a zero-training point cloud classification network based on the recently introduced BLIP-2 visual-language model. PointBLIP is adept at processing similarities between multi-images and multi-prompts. We separately introduce a novel method for point cloud zero-shot and few-shot classification, which involves comparing multiple features to achieve effective classification. Simultaneously, we enhance the input data quality for both the image and text sides of PointBLIP. In point cloud zero-shot classification tasks, we outperform state-of-the-art methods on three benchmark datasets. For few-shot classification tasks, to the best of our knowledge, we present the first zero-training few-shot point cloud method, surpassing previous works under the same conditions and showcasing comparable performance to full-training methods.

Despite the importance of few-shot learning, the lack of labeled training data in the real world, makes it extremely challenging for existing machine learning methods as this limited data set does not represent the data variance well. In this research, we suggest employing a generative approach using variational autoencoders (VAEs), which can be used specifically to optimize few-shot learning tasks by generating new samples with more intra-class variations. The purpose of our research is to increase the size of the training data set using various methods to improve the accuracy and robustness of the few-shot face recognition. Specifically, we employ the VAE generator to increase the size of the training data set, including the basic and the novel sets while utilizing transfer learning as the backend. Based on extensive experimental research, we analyze various data augmentation methods to observe how each method affects the accuracy of face recognition. We conclude that the face generation method we proposed can effectively improve the recognition accuracy rate to 96.47% using both the base and the novel sets.

MoS₂/TiO₂ nanostructures made of MoS₂ nanoparticles covering TiO₂ nanosheets have been synthesized, either via ex-situ or in-situ approaches. Morphology and structure of MoS₂/TiO₂ hybrid nanostructures have been investigated and imaged by means of X-ray diffraction (XRD) analysis and high-resolution transmission electron microscopy (HRTEM), while the vibrational and the optical properties have been investigated by Raman, Fourier-transform infrared spectroscopy (FTIR) and UV−visible (UV-Vis) techniques. The different stacking degrees together with the size distribution of the MoS₂ nanosheets, decorating the TiO₂ nanosheets, have been carefully obtained from HRTEM images. The nature of the surface sites on the main exposed faces of both materials has been detected by means of in-situ FTIR spectra of adsorbed CO probe molecule. The results coming from the ex-situ and in-situ approaches will be compared, by highlighting the role of the synthesis processes in affecting morphology and structure of MoS₂ nanosheets, including their curvature, surface defects, and stacking order. Some more, it will be shown that the in-situ approach is affecting the reactivity of the TiO₂ nanosheets too, hence in turn affects the MoS₂/TiO₂ nanosheets interaction.

This work presents first time successfully fabricated silica few-mode microstructured optical fiber (MOF) with hollow GeO2-doped ring core and strongly induced twisting up to 790 revolutions per meter. Some technological issues for manufacturing of GeO2-doped supporting elements for large hollow cores as well as described above complicated spun MOFs are discussed. We introduce some results of tests, performed for pilot samples of designed and manufactured described above untwisted and twisted MOFs with outer diameter 65 µm and hollow ring core inner diameter 30.5 µm under wall thickness 1.7 µm and refractive index difference n=0.030, including their geometrical parameters, basic transmission characteristics and measurements of far-field laser beam profile patterns.

This paper discusses approaches to the numerical integration of the coupled nonlinear Schrödinger equations system in case of few-mode wave propagation. The wave propagation assumes the propagation of up to nine modes of light in an optical fiber. In this case, the light propagation is described by the non-linear coupled Schrödinger equation system, where propagation of each mode is described by own Schrödinger equation with other modes interactions. In this case, the non-linear coupled Schrödinger equation system solving becomes increasingly complex, because each mode affects the propagation of other modes. The suggested solution is based on the direct numerical integration approach, which is based on a finite-difference integration scheme. The well-known explicit finite-difference integration scheme approach fails, due to the non-stability of the computing scheme. Due to this fact, the combined explicit/implicit finite-difference integration scheme, based on the implicit Crank–Nicolson finite-difference scheme, is used. It allows ensuring the stability of the computing scheme. Moreover, this approach allows separating the whole equation system on the independent equation system for each wave mode at each integration step. Additionally, the algorithm of numerical solution refining at each step and the integration method with automatic integration step selection are used. The suggested approach has performance gains (or resolutions) up to three or more orders of magnitude in comparison with the split-step Fourier method due to the fact that there is no need to produce direct and inverse Fourier transforms at each integration step. The main advantage of the proposed method is the ability to calculate the propagation of an arbitrary number of modes in the fiber.