Preferentially selective extraction of lithium from spent LiCoO_(2)cathodes by medium-temperature carbon reduction roasting

Lithium recovery from spent lithium-ion batteries(LIBs)have attracted extensive attention due to the skyrocketing price of lithium.The medium-temperature carbon reduction roasting was proposed to preferential selective extraction of lithium from spent Li-CoO_(2)(LCO)cathodes to overcome the incomplete recovery and loss of lithium during the recycling process.The LCO layered structure was destroyed and lithium was completely converted into water-soluble Li2CO_(3)under a suitable temperature to control the reduced state of the cobalt oxide.The Co metal agglomerates generated during medium-temperature carbon reduction roasting were broken by wet grinding and ultrasonic crushing to release the entrained lithium.The results showed that 99.10%of the whole lithium could be recovered as Li2CO_(3)with a purity of 99.55%.This work provided a new perspective on the preferentially selective extraction of lithium from spent lithium batteries.

Automatic Extraction Method of 3D Feature Guidelines for Complex Cultural Relic Surfaces Based on Point Cloud

Cultural relics line graphic serves as a crucial form of traditional artifact information documentation,which is a simple and intuitive product with low cost of displaying compared with 3D models.Dimensionality reduction is undoubtedly necessary for line drawings.However,most existing methods for artifact drawing rely on the principles of orthographic projection that always cannot avoid angle occlusion and data overlapping while the surface of cultural relics is complex.Therefore,conformal mapping was introduced as a dimensionality reduction way to compensate for the limitation of orthographic projection.Based on the given criteria for assessing surface complexity,this paper proposed a three-dimensional feature guideline extraction method for complex cultural relic surfaces.A 2D and 3D combined factor that measured the importance of points on describing surface features,vertex weight,was designed.Then the selection threshold for feature guideline extraction was determined based on the differences between vertex weight and shape index distributions.The feasibility and stability were verified through experiments conducted on real cultural relic surface data.Results demonstrated the ability of the method to address the challenges associated with the automatic generation of line drawings for complex surfaces.The extraction method and the obtained results will be useful for line graphic drawing,displaying and propaganda of cultural relics.

Effects of conductive agent type on lithium extraction from salt lake brine with LiFePO_(4) electrodes

Electrochemical lithium extraction from salt lakes is an effective strategy for obtaining lithium at a low cost.Nevertheless,the elevated Mg:Li ratio and the presence of numerous coexisting ions in salt lake brines give rise to challenges,such as prolonged lithium extraction periods,diminished lithium extraction efficiency,and considerable environmental pollution.In this work,Li FePO4(LFP)served as the electrode material for electrochemical lithium extraction.The conductive network in the LFP electrode was optimized by adjusting the type of conductive agent.This approach resulted in high lithium extraction efficiency and extended cycle life.When the single conductive agent of acetylene black(AB)or multiwalled carbon nanotubes(MWCNTs)was replaced with the mixed conductive agent of AB/MWCNTs,the average diffusion coefficient of Li+in the electrode increased from 2.35×10^(-9)or 1.77×10^(-9)to 4.21×10^(-9)cm^(2)·s^(-1).At the current density of 20 mA·g^(-1),the average lithium extraction capacity per gram of LFP electrode increased from 30.36 mg with the single conductive agent(AB)to 35.62 mg with the mixed conductive agent(AB/MWCNTs).When the mixed conductive agent was used,the capacity retention of the electrode after 30 cycles reached 82.9%,which was considerably higher than the capacity retention of 65.8%obtained when the single AB was utilized.Meanwhile,the electrode with mixed conductive agent of AB/MWCNTs provided good cycling performance.When the conductive agent content decreased or the loading capacity increased,the electrode containing the mixed conductive agent continued to show excellent electrochemical performance.Furthermore,a self-designed,highly efficient,continuous lithium extraction device was constructed.The electrode utilizing the AB/MWCNT mixed conductive agent maintained excellent adsorption capacity and cycling performance in this device.This work provides a new perspective for the electrochemical extraction of lithium using LFP electrodes.

Research advances in enhanced coal seam gas extraction by controllable shock wave fracturing

With the continuous increase of mining in depth,the gas extraction faces the challenges of low permeability,great ground stress,high temperature and large gas pressure in coal seam.The controllable shock wave(CSW),as a new method for enhancing permeability of coal seam to improve gas extraction,features in the advantages of high efficiency,eco-friendly,and low cost.In order to better utilize the CSW into gas extraction in coal mine,the mechanism and feasibility of CSW enhanced extraction need to be studied.In this paper,the basic principles,the experimental tests,the mathematical models,and the on-site tests of CSW fracturing coal seams are reviewed,thereby its future research directions are provided.Based on the different media between electrodes,the CSW can be divided into three categories:hydraulic effect,wire explosion and excitation of energetic materials by detonating wire.During the process of propagation and attenuation of the high-energy shock wave in coal,the shock wave and bubble pulsation work together to produce an enhanced permeability effect on the coal seam.The stronger the strength of the CSW is,the more cracks created in the coal is,and the greater the length,width and area of the cracks being.The repeated shock on the coal seam is conducive to the formation of complex network fracture system as well as the reduction of coal seam strength,but excessive shock frequency will also damage the coal structure,resulting in the limited effect of the enhanced gas extraction.Under the influence of ground stress,the crack propagation in coal seam will be restrained.The difference of horizontal principal stress has a significant impact on the shape,propagation direction and connectivity of the CSW induced cracks.The permeability enhancement effect of CSW is affected by the breakage degree of coal seam.The shock wave is absorbed by the broken coal,which may hinder the propagation of CSW,resulting in a poor effect of permeability enhancement.When arranging two adjacent boreholes for CSW permeability enhancement test,the spacing of boreholes should not be too close,which may lead to negative pressure mutual pulling in the early stage of drainage.At present,the accurate method for effectively predicting the CSW permeability enhanced range should be further investigated.

Insight into the experiment and extraction mechanism for separating carbazole from anthracene oil with quaternary ammonium-based deep eutectic solvents

Carbazole is an irreplaceable basic organic chemical raw material and intermediate in industry.The separation of carbazole from anthracene oil by environmental benign solvents is important but still a challenge in chemical engineering.Deep eutectic solvents (DESs) as a sustainable green separation solvent have been proposed for the separation of carbazole from model anthracene oil.In this research,three quaternary ammonium-based DESs were prepared using ethylene glycol (EG) as hydrogen bond donor and tetrabutylammonium chloride (TBAC),tetrabutylammonium bromide or choline chloride as hydrogen bond acceptors.To explore their extraction performance of carbazole,the conductor-like screening model for real solvents (COSMO-RS) model was used to predict the activity coefficient at infinite dilution (γ^(∞)) of carbazole in DESs,and the result indicated TBAC:EG (1:2) had the stronger extraction ability for carbazole due to the higher capacity at infinite dilution (C^(∞)) value.Then,the separation performance of these three DESs was evaluated by experiments,and the experimental results were in good agreement with the COSMO-RS prediction results.The TBAC:EG (1:2) was determined as the most promising solvent.Additionally,the extraction conditions of TBAC:EG (1:2) were optimized,and the extraction efficiency,distribution coefficient and selectivity of carbazole could reach up to 85.74%,30.18 and 66.10%,respectively.Moreover,the TBAC:EG (1:2) could be recycled by using environmentally friendly water as antisolvent.In addition,the separation performance of TBAC:EG (1:2) was also evaluated by real crude anthracene,the carbazole was obtained with purity and yield of 85.32%,60.27%,respectively.Lastly,the extraction mechanism was elucidated byσ-profiles and interaction energy analysis.Theoretical calculation results showed that the main driving force for the extraction process was the hydrogen bonding ((N–H...Cl) and van der Waals interactions (C–H...O and C–H...π),which corresponding to the blue and green isosurfaces in IGMH analysis.This work presented a novel method for separating carbazole from crude anthracene oil,and will provide an important reference for the separation of other high value-added products from coal tar.

Experimental investigation on coal pore-fracture variation and fractal characteristics synergistically affected by solvents for improving clean gas extraction

Chemical solvents instead of pure water being as hydraulic fracturing fluid could effectively increase permeability and improve clean methane extraction efficiency.However,pore-fracture variation features of lean coal synergistically affected by solvents have not been fully understood.Ultrasonic testing,nuclear magnetic resonance analysis,liquid phase mass spectrometry was adopted to comprehensively analyze pore-fracture change characteristics of lean coal treated by combined solvent(NMP and CS_(2)).Meanwhile,quantitative characterization of above changing properties was conducted using geometric fractal theory.Relationship model between permeability,fractal dimension and porosity were established.Results indicate that the end face fractures of coal are well developed after CS2and combined solvent treatments,of which,end face box-counting fractal dimensions range from 1.1227 to 1.4767.Maximum decreases in ultrasonic longitudinal wave velocity of coal affected by NMP,CS_(2)and combined solvent are 2.700%,20.521%,22.454%,respectively.Solvent treatments could lead to increasing amount of both mesopores and macropores.Decrease ratio of fractal dimension Dsis 0.259%–2.159%,while permeability increases ratio of NMR ranges from 0.1904 to 6.4486.Meanwhile,combined solvent could dissolve coal polar and non-polar small molecules and expand flow space.Results could provide reference for solvent selection and parameter optimization of permeability-enhancement technology.

Design method of extractant for liquid-liquid extraction based on elements and chemical bonds

In the petrochemical industry process, the relative volatility between the components to be separated is close to one or the azeotrope that systems are difficult to separate. Liquid-liquid extraction is a common and effective separation method, and selecting an extraction agent is the key to extraction technology research. In this paper, a design method of extractants based on elements and chemical bonds was proposed. A knowledge-based molecular design method was adopted to pre-select elements and chemical bond groups. The molecules were automatically synthesized according to specific combination rules to avoid the problem of “combination explosion” of molecules. The target properties of the extractant were set, and the extractant meeting the requirements was selected by predicting the correlation physical properties of the generated molecules. Based on the separation performance of the extractant in liquid-liquid extraction and the relative importance of each index, the fuzzy comprehensive evaluation membership function was established, the analytic hierarchy process determined the mass ratio of each index, and the consistency test results were passed. The results of case study based on quantum chemical analysis demonstrated that effective determination of extractants for the analysis of benzene-cyclohexane systems. The results unanimously prove that the method has important theoretical significance and application value.

A Joint Entity Relation Extraction Model Based on Relation Semantic Template Automatically Constructed

The joint entity relation extraction model which integrates the semantic information of relation is favored by relevant researchers because of its effectiveness in solving the overlapping of entities,and the method of defining the semantic template of relation manually is particularly prominent in the extraction effect because it can obtain the deep semantic information of relation.However,this method has some problems,such as relying on expert experience and poor portability.Inspired by the rule-based entity relation extraction method,this paper proposes a joint entity relation extraction model based on a relation semantic template automatically constructed,which is abbreviated as RSTAC.This model refines the extraction rules of relation semantic templates from relation corpus through dependency parsing and realizes the automatic construction of relation semantic templates.Based on the relation semantic template,the process of relation classification and triplet extraction is constrained,and finally,the entity relation triplet is obtained.The experimental results on the three major Chinese datasets of DuIE,SanWen,and FinRE showthat the RSTAC model successfully obtains rich deep semantics of relation,improves the extraction effect of entity relation triples,and the F1 scores are increased by an average of 0.96% compared with classical joint extraction models such as CasRel,TPLinker,and RFBFN.

Phase equilibrium data prediction and process optimizationin butadiene extraction process

In response to the lack of reliable physical parameters in the process simulation of the butadiene extraction,a large amount of phase equilibrium data were collected in the context of the actual process of butadiene production by acetonitrile.The accuracy of five prediction methods,UNIFAC(UNIQUAC Functional-group Activity Coefficients),UNIFAC-LL,UNIFAC-LBY,UNIFAC-DMD and COSMO-RS,applied to the butadiene extraction process was verified using partial phase equilibrium data.The results showed that the UNIFAC-DMD method had the highest accuracy in predicting phase equilibrium data for the missing system.COSMO-RS-predicted multiple systems showed good accuracy,and a large number of missing phase equilibrium data were estimated using the UNIFAC-DMD method and COSMO-RS method.The predicted phase equilibrium data were checked for consistency.The NRTL-RK(non-Random Two Liquid-Redlich-Kwong Equation of State)and UNIQUAC thermodynamic models were used to correlate the phase equilibrium data.Industrial device simulations were used to verify the accuracy of the thermodynamic model applied to the butadiene extraction process.The simulation results showed that the average deviations of the simulated results using the correlated thermodynamic model from the actual values were less than 2%compared to that using the commercial simulation software,Aspen Plus and its database.The average deviation was much smaller than that of the simulations using the Aspen Plus database(>10%),indicating that the obtained phase equilibrium data are highly accurate and reliable.The best phase equilibrium data and thermodynamic model parameters for butadiene extraction are provided.This improves the accuracy and reliability of the design,optimization and control of the process,and provides a basis and guarantee for developing a more environmentally friendly and economical butadiene extraction process.

Highly selective extraction of aromatics from aliphatics by using metal chloride-based ionic liquids

The separation of aromatics from aliphatics is essential for achieving maximum exploitation of oil resources in the petrochemical industry.In this study,a series of metal chloride-based ionic liquids were prepared and their performances in the separation of 1,2,3,4-tetrahydronaphthalene(tetralin)/dodecane and tetralin/decalin systems were studied.Among these ionic liquids,1-ethyl-3-methylimidazolium tetrachloroferrate([EMIM][FeCl_(4)])with the highest selectivity was used as the extractant.Density functional theory calculations showed that[EMIM][FeCl_(4)]interacted more strongly with tetralin than with dodecane and decalin.Energy decomposition analysis of[EMIM][FeCl_(4)]-tetralin indicated that electrostatics and dispersion played essential roles,and induction cannot be neglected.The van der Waals forces was a main effect in[EMIM][FeCl_(4)]-tetralin by independent gradient model analysis.The tetralin distribution coefficient and selectivity were 0.8 and 110,respectively,with 10%(mol)tetralin in the initial tetralin/dodecane system,and 0.67 and 19.5,respectively,with 10%(mol)tetralin in the initial tetralin/decalin system.The selectivity increased with decreasing alkyl chain length of the extractant.The influence of the extraction temperature,extractant dosage,and initial concentrations of the system components on the separation performance were studied.Recycling experiments showed that the regenerated[EMIM][FeCl_(4)]could be used repeatedly.

Influence of extraction voltage on electron and ion behavior characteristics

The characteristics of the extracted ion current have a significant impact on the design and testing of ion source performance.In this paper,a 2D in space and 3D in velocity space particle in cell(2D3V PIC)method is utilized to simulate plasma motion and ion extraction characteristics under various initial plasma velocity distributions and extraction voltages in a Cartesian coordinate system.The plasma density is of the order of 10^(15)m^(-3)-10^(16)m^(-3)and the extraction voltage is of the order of 100 V-1000 V.The study investigates the impact of various extraction voltages on the velocity and density distributions of electrons and positive ions,and analyzes the influence of different initial plasma velocity distributions on the extraction current.The simulation results reveal that the main reason for the variation of extraction current is the spacecharge force formed by the relative aggregation of positive and negative net charges.This lays the foundation for a deeper understanding of extraction beam characteristics.

Upper-rim-modified thioether thiacalix[4]arene used for liquid-liquid extraction of Au(Ⅲ)and Pd(Ⅱ)in hydrochloric acid medium

Gold(Au)and palladium(Pd)play an increasing role in the production and human life;Therefore,it is of great significance to study their recovery.A 5,11,17,23-tetra-ethylthio-25,26,27,28-tetra-hydroxyl thiacalix[4]arene(TCAET)was synthesized specifically for the capture of Au(Ⅲ)and Pd(Ⅱ)from HCl medium by liquid-liquid extraction.In a 0.1 mol·L^(-1)HCl medium,the transfer of Au(Ⅲ)and Pd(Ⅱ)from the aqueous phase to the organic phase was highly efficient,with a transfer ratio of 100%for Au(Ⅲ)and 98%for Pd(Ⅱ).Furthermore,the extraction equilibrium time for Au(Ⅲ)was just 5 min.Job's method data demonstrated that TCAET formed complexes with Au(Ⅲ)and Pd(Ⅱ)in a ratio of 2:3 and 1:1,respectively,during the extraction process.TCAET showed high selectivity toward Pd(Ⅱ)and Au(Ⅲ)over other competing metal ions.Moreover,both Au(Ⅲ)and Pd(Ⅱ)could be successfully stripped from the loaded organic phases with a 1.0 mol·L^(-1)thiourea in 0.5 mol·L^(-1)HCl and 0.5 mol·L^(-1)thiourea in 0.5 mol·L^(-1)HCl,respectively.Results obtained from five consecutive extraction-stripping cycles showed good reusability of TCAET toward Au(Ⅲ)and Pd(Ⅱ)recovery.The conclusion can provide a certain reference for thiacalixarene in the recovery of precious metal species.

Anomaly Detection in Imbalanced Encrypted Traffic with Few Packet Metadata-Based Feature Extraction

In the IoT(Internet of Things)domain,the increased use of encryption protocols such as SSL/TLS,VPN(Virtual Private Network),and Tor has led to a rise in attacks leveraging encrypted traffic.While research on anomaly detection using AI(Artificial Intelligence)is actively progressing,the encrypted nature of the data poses challenges for labeling,resulting in data imbalance and biased feature extraction toward specific nodes.This study proposes a reconstruction error-based anomaly detection method using an autoencoder(AE)that utilizes packet metadata excluding specific node information.The proposed method omits biased packet metadata such as IP and Port and trains the detection model using only normal data,leveraging a small amount of packet metadata.This makes it well-suited for direct application in IoT environments due to its low resource consumption.In experiments comparing feature extraction methods for AE-based anomaly detection,we found that using flowbased features significantly improves accuracy,precision,F1 score,and AUC(Area Under the Receiver Operating Characteristic Curve)score compared to packet-based features.Additionally,for flow-based features,the proposed method showed a 30.17%increase in F1 score and improved false positive rates compared to Isolation Forest and OneClassSVM.Furthermore,the proposedmethod demonstrated a 32.43%higherAUCwhen using packet features and a 111.39%higher AUC when using flow features,compared to previously proposed oversampling methods.This study highlights the impact of feature extraction methods on attack detection in imbalanced,encrypted traffic environments and emphasizes that the one-class method using AE is more effective for attack detection and reducing false positives compared to traditional oversampling methods.

CMMCAN:Lightweight Feature Extraction and Matching Network for Endoscopic Images Based on Adaptive Attention

In minimally invasive surgery,endoscopes or laparoscopes equipped with miniature cameras and tools are used to enter the human body for therapeutic purposes through small incisions or natural cavities.However,in clinical operating environments,endoscopic images often suffer from challenges such as low texture,uneven illumination,and non-rigid structures,which affect feature observation and extraction.This can severely impact surgical navigation or clinical diagnosis due to missing feature points in endoscopic images,leading to treatment and postoperative recovery issues for patients.To address these challenges,this paper introduces,for the first time,a Cross-Channel Multi-Modal Adaptive Spatial Feature Fusion(ASFF)module based on the lightweight architecture of EfficientViT.Additionally,a novel lightweight feature extraction and matching network based on attention mechanism is proposed.This network dynamically adjusts attention weights for cross-modal information from grayscale images and optical flow images through a dual-branch Siamese network.It extracts static and dynamic information features ranging from low-level to high-level,and from local to global,ensuring robust feature extraction across different widths,noise levels,and blur scenarios.Global and local matching are performed through a multi-level cascaded attention mechanism,with cross-channel attention introduced to simultaneously extract low-level and high-level features.Extensive ablation experiments and comparative studies are conducted on the HyperKvasir,EAD,M2caiSeg,CVC-ClinicDB,and UCL synthetic datasets.Experimental results demonstrate that the proposed network improves upon the baseline EfficientViT-B3 model by 75.4%in accuracy(Acc),while also enhancing runtime performance and storage efficiency.When compared with the complex DenseDescriptor feature extraction network,the difference in Acc is less than 7.22%,and IoU calculation results on specific datasets outperform complex dense models.Furthermore,this method increases the F1 score by 33.2%and accelerates runtime by 70.2%.It is noteworthy that the speed of CMMCAN surpasses that of comparative lightweight models,with feature extraction and matching performance comparable to existing complex models but with faster speed and higher cost-effectiveness.

Network Configuration Entity Extraction Method Based on Transformer with Multi-Head Attention Mechanism

Nowadays,ensuring thequality of networkserviceshas become increasingly vital.Experts are turning toknowledge graph technology,with a significant emphasis on entity extraction in the identification of device configurations.This research paper presents a novel entity extraction method that leverages a combination of active learning and attention mechanisms.Initially,an improved active learning approach is employed to select the most valuable unlabeled samples,which are subsequently submitted for expert labeling.This approach successfully addresses the problems of isolated points and sample redundancy within the network configuration sample set.Then the labeled samples are utilized to train the model for network configuration entity extraction.Furthermore,the multi-head self-attention of the transformer model is enhanced by introducing the Adaptive Weighting method based on the Laplace mixture distribution.This enhancement enables the transformer model to dynamically adapt its focus to words in various positions,displaying exceptional adaptability to abnormal data and further elevating the accuracy of the proposed model.Through comparisons with Random Sampling(RANDOM),Maximum Normalized Log-Probability(MNLP),Least Confidence(LC),Token Entrop(TE),and Entropy Query by Bagging(EQB),the proposed method,Entropy Query by Bagging and Maximum Influence Active Learning(EQBMIAL),achieves comparable performance with only 40% of the samples on both datasets,while other algorithms require 50% of the samples.Furthermore,the entity extraction algorithm with the Adaptive Weighted Multi-head Attention mechanism(AW-MHA)is compared with BILSTM-CRF,Mutil_Attention-Bilstm-Crf,Deep_Neural_Model_NER and BERT_Transformer,achieving precision rates of 75.98% and 98.32% on the two datasets,respectively.Statistical tests demonstrate the statistical significance and effectiveness of the proposed algorithms in this paper.

Feature extraction for machine learning-based intrusion detection in IoT networks

A large number of network security breaches in IoT networks have demonstrated the unreliability of current Network Intrusion Detection Systems(NIDSs).Consequently,network interruptions and loss of sensitive data have occurred,which led to an active research area for improving NIDS technologies.In an analysis of related works,it was observed that most researchers aim to obtain better classification results by using a set of untried combinations of Feature Reduction(FR)and Machine Learning(ML)techniques on NIDS datasets.However,these datasets are different in feature sets,attack types,and network design.Therefore,this paper aims to discover whether these techniques can be generalised across various datasets.Six ML models are utilised:a Deep Feed Forward(DFF),Convolutional Neural Network(CNN),Recurrent Neural Network(RNN),Decision Tree(DT),Logistic Regression(LR),and Naive Bayes(NB).The accuracy of three Feature Extraction(FE)algorithms is detected;Principal Component Analysis(PCA),Auto-encoder(AE),and Linear Discriminant Analysis(LDA),are evaluated using three benchmark datasets:UNSW-NB15,ToN-IoT and CSE-CIC-IDS2018.Although PCA and AE algorithms have been widely used,the determination of their optimal number of extracted dimensions has been overlooked.The results indicate that no clear FE method or ML model can achieve the best scores for all datasets.The optimal number of extracted dimensions has been identified for each dataset,and LDA degrades the performance of the ML models on two datasets.The variance is used to analyse the extracted dimensions of LDA and PCA.Finally,this paper concludes that the choice of datasets significantly alters the performance of the applied techniques.We believe that a universal(benchmark)feature set is needed to facilitate further advancement and progress of research in this field.

Online identification and extraction method of regional large-scale adjustable load-aggregation characteristics

This article introduces the concept of load aggregation,which involves a comprehensive analysis of loads to acquire their external characteristics for the purpose of modeling and analyzing power systems.The online identification method is a computer-involved approach for data collection,processing,and system identification,commonly used for adaptive control and prediction.This paper proposes a method for dynamically aggregating large-scale adjustable loads to support high proportions of new energy integration,aiming to study the aggregation characteristics of regional large-scale adjustable loads using online identification techniques and feature extraction methods.The experiment selected 300 central air conditioners as the research subject and analyzed their regulation characteristics,economic efficiency,and comfort.The experimental results show that as the adjustment time of the air conditioner increases from 5 minutes to 35 minutes,the stable adjustment quantity during the adjustment period decreases from 28.46 to 3.57,indicating that air conditioning loads can be controlled over a long period and have better adjustment effects in the short term.Overall,the experimental results of this paper demonstrate that analyzing the aggregation characteristics of regional large-scale adjustable loads using online identification techniques and feature extraction algorithms is effective.

Weakly Supervised Network with Scribble-Supervised and Edge-Mask for Road Extraction from High-Resolution Remote Sensing Images

Significant advancements have been achieved in road surface extraction based on high-resolution remote sensingimage processing. Most current methods rely on fully supervised learning, which necessitates enormous humaneffort to label the image. Within this field, other research endeavors utilize weakly supervised methods. Theseapproaches aim to reduce the expenses associated with annotation by leveraging sparsely annotated data, such asscribbles. This paper presents a novel technique called a weakly supervised network using scribble-supervised andedge-mask (WSSE-net). This network is a three-branch network architecture, whereby each branch is equippedwith a distinct decoder module dedicated to road extraction tasks. One of the branches is dedicated to generatingedge masks using edge detection algorithms and optimizing road edge details. The other two branches supervise themodel’s training by employing scribble labels and spreading scribble information throughout the image. To addressthe historical flaw that created pseudo-labels that are not updated with network training, we use mixup to blendprediction results dynamically and continually update new pseudo-labels to steer network training. Our solutiondemonstrates efficient operation by simultaneously considering both edge-mask aid and dynamic pseudo-labelsupport. The studies are conducted on three separate road datasets, which consist primarily of high-resolutionremote-sensing satellite photos and drone images. The experimental findings suggest that our methodologyperforms better than advanced scribble-supervised approaches and specific traditional fully supervised methods.

Enhancing Relational Triple Extraction in Specific Domains:Semantic Enhancement and Synergy of Large Language Models and Small Pre-Trained Language Models

In the process of constructing domain-specific knowledge graphs,the task of relational triple extraction plays a critical role in transforming unstructured text into structured information.Existing relational triple extraction models facemultiple challenges when processing domain-specific data,including insufficient utilization of semantic interaction information between entities and relations,difficulties in handling challenging samples,and the scarcity of domain-specific datasets.To address these issues,our study introduces three innovative components:Relation semantic enhancement,data augmentation,and a voting strategy,all designed to significantly improve the model’s performance in tackling domain-specific relational triple extraction tasks.We first propose an innovative attention interaction module.This method significantly enhances the semantic interaction capabilities between entities and relations by integrating semantic information fromrelation labels.Second,we propose a voting strategy that effectively combines the strengths of large languagemodels(LLMs)and fine-tuned small pre-trained language models(SLMs)to reevaluate challenging samples,thereby improving the model’s adaptability in specific domains.Additionally,we explore the use of LLMs for data augmentation,aiming to generate domain-specific datasets to alleviate the scarcity of domain data.Experiments conducted on three domain-specific datasets demonstrate that our model outperforms existing comparative models in several aspects,with F1 scores exceeding the State of the Art models by 2%,1.6%,and 0.6%,respectively,validating the effectiveness and generalizability of our approach.

ConvNeXt-UperNet-Based Deep Learning Model for Road Extraction from High-Resolution Remote Sensing Images

When existing deep learning models are used for road extraction tasks from high-resolution images,they are easily affected by noise factors such as tree and building occlusion and complex backgrounds,resulting in incomplete road extraction and low accuracy.We propose the introduction of spatial and channel attention modules to the convolutional neural network ConvNeXt.Then,ConvNeXt is used as the backbone network,which cooperates with the perceptual analysis network UPerNet,retains the detection head of the semantic segmentation,and builds a new model ConvNeXt-UPerNet to suppress noise interference.Training on the open-source DeepGlobe and CHN6-CUG datasets and introducing the DiceLoss on the basis of CrossEntropyLoss solves the problem of positive and negative sample imbalance.Experimental results show that the new network model can achieve the following performance on the DeepGlobe dataset:79.40%for precision(Pre),97.93% for accuracy(Acc),69.28% for intersection over union(IoU),and 83.56% for mean intersection over union(MIoU).On the CHN6-CUG dataset,the model achieves the respective values of 78.17%for Pre,97.63%for Acc,65.4% for IoU,and 81.46% for MIoU.Compared with other network models,the fused ConvNeXt-UPerNet model can extract road information better when faced with the influence of noise contained in high-resolution remote sensing images.It also achieves multiscale image feature information with unified perception,ultimately improving the generalization ability of deep learning technology in extracting complex roads from high-resolution remote sensing images.