Multimodal sentiment analysis, an inherently challenging research area, integrates textual, audio, and visual modalities to analyze human emotional tendencies. Existing studies suggest that the textual modality plays a dominant role in sentiment prediction. However, this predominance raises a potential issue: during training, machine learning models tend to learn spurious correlations between the input and the output, leading to an overreliance on textual information. This overreliance may cause models to incorrectly model spurious correlations between textual and sentiment labels, thus undermining the model's generalization ability. To address this challenge, an innovative counterfactual text debiasing(CFTB) algorithm is proposed for multimodal sentiment analysis. Our framework first employs causal graph to thoroughly analyze the causal relationships among the three modalities and the sentiment labels.Then, an auxiliary textual model is designed to precisely quantify the direct impacts of the textual modality to sentiment prediction and leverage an attention mechanism to accurately capture textual features that might introduce spurious correlations. During the inference phase, the CFTB algorithm demonstrates its unique advantage: it intelligently isolates the negative impacts caused by spurious textual associations from the overall multimodal information, while retaining and enhancing the beneficial information within the textual modality that genuinely contributes to sentiment prediction. Experiments on the MOSEI and MOSI datasets show that this framework can be effectively integrated into existing methods and has good generalization performance.

Aiming at the difficulties present in current automatic sleep staging methods, a method for automatic sleep staging of EEG and ECG dual modal signals by combining U²-Net and CBAM fusion attention is proposed.Firstly, the EEG-ECG signals in the MIT-BIH public dataset used in this paper are preprocessed. Then, the U²-Net network with multi-scale feature extraction module is used to extract waveform features in EEG and ECG in parallel. Secondly, CBAM fusion attention is used to assign weights to all features. Finally, the Softmax activation function is used to classify sleep periods into six. The results show that when sleep staging is performed based on U²-Net and CBAM fusion attention models, the overall accuracy of hexaclassification using ECG single-modal signals is 80.2%, and the F1 score is 75.3%. The overall accuracy of six classifications using EEG single-modal signals was 85.8%, and the F1 score was 81.7%;The overall accuracy of the six classifications using EEG-ECG dual-modal signals was 90.4%, and the F1 score was 85.6%. This shows that the bimodal sleep staging model proposed in this paper is feasible and effective, and provides a new idea for automatic sleep staging.

Imaging genetics suggests that there is a certain degree of correlation between neuroimaging and genes, leading researchers to pay attention to the analysis of diseases using genetic variations and imaging data. In practice, clinical doctors usually have limited data availability but still aspire to employ deep learning method for real-world problems. Considering the expanding data scale and expensive annotation costs, it becomes essential to develop an unsupervised learning method capable of utilizing multimodal data. To meet these needs, a representation learning method based on multimodal tabular data with contrastive learning (MTCL) is proposed. The model leverages resting-state functional magnetic resonance imaging (rs-fMRI) and single nucleotide polymorphisms (SNP) data without requiring any labeled information. To enhance interpretability, the model first transforms rs-fMRI and SNP data into a tabular structure through a feature extraction module. Then, a multimodal tabular data contrastive learning method is employed to fuse the dataset and obtain the fused data representation. On the dataset of major depressive disorder (MDD), our proposed method effectively improves the diagnostic performance of MDD. Additionally, the MTCL method combines model attribution techniques to explore imaging and genetic biomarkers associated with MDD, enhancing the interpretability of the model and aiding researchers in understanding the mechanisms underlying the disease.

Though online reviews on social media platforms have been widely used in tourism research as data analytical sources in recent years, how they can be applied to destination recommendation needs further investigation.The review data of 239 5A scenic spots in China was retrieved from ctrip.com by Python programming and web-crawling technology. Natural language processing and deep learning technologies including BERT (Bidirectional Encoder Representations from Transformers) and word embedding were then imported to build a destination recommendation system for tourist destinations. The model was trained and validated on a dataset containing 57 360 reviews, with a classification accuracy of around 78% reached on 14 340 pieces of test data. Experimental results show that, with the aid of other tourists’ travel experiences and image perception, the proposed model can facilitate potential tourists in finding their ideal destinations to explore the first step of itinerary planning. The findings of this study extend the research scope of online reviews within tourism and hospitality and provide new insights into pre-trip travel counseling.

To clarify the dynamic changes and clinical significance of heart rate variability in adolescent depression with aerobic exercise therapy,60 students of mild to moderate depression enrolled in 2022 were divided into the study group (5 cases dropped, 25 cases completed ) and the control group (2 cases dropped, 28 cases completed) equally. The patients in the study group received aerobic exercise therapy for 12 weeks, the patients in the control group did not receive any treatment.The clinical efficacy and heart rate variability index between the two groups were compared.The clinical efficacy and cure rate of the study group were significantly higher than those of the control group at the end of the 12th week (P<0.05).The 17-item Hamilton Depression Scale (HAMD-17) score in the two groups was significantly lower than that at baseline (P<0.01).Compared with the control group, the study group showed a more significant decrease in HAMD-17 at the end of the 12th week (P<0.001). The standard deviation of NN intervals(SDNN),standard deviation average of NN intervals(SDANN), and root mean square of successive RR interval differences(RMSSD) in the study group at the end of the 12th week were significantly increased than those before treatment (P<0.001).Compared with the control group, the scores of SDNN, SDANN, and RMSSD in the study group at the end of the 12th week changed more significantly (P<0.001). Negative correlations were found between baseline HAMD-17 score and SDNN, SDANN, and RMSSD (r=-0.343, P=0.012; r=-0.328, P=0.017; r=-0.401, P=0.003; respectively). The results indicate that 12-week aerobic exercise can significantly improve depression symptoms and autonomic nervous system function in adolescent depression patients. The severity of depression is related to impaired autonomic nervous system function, and heart rate variability can become a biological indicator for evaluating the efficacy of aerobic exercise in treating depression.

This study introduces a technique known as large-field optical sectioning structured illumination microscopy(large-field optical sectioning structured illumination microscopy, LF-OS-SIM) designed for three-dimensional tomographic microscopic imaging of thick specimens. The technique utilizes a one-dimensional grating projection to generate a striped structured light field, in conjunction with a spatial light modulator (spatial light modulator, SLM) for fast phase-shifting of the structured light spectrum. Compared to conventional OS-SIM techniques that are based on SLM/DMD projection, LF-OS-SIM enhances the spatial bandwidth product (spatial bandwidth product, SBP) by 4.7 times. Additionally, digital phase-shifting using the SLM is implemented to achieve a slicing speed of 20 frames per second. The LF-OS-SIM was employed for three-dimensional tomographic microscopic imaging of objects such as coins, three-dimensionally distributed fluorescent beads, and biological specimens. The results indicated that the imaging field of view (FOV) for LS-OS-SIM reached 1 030×780 μm³, with an axial tomographic imaging accuracy of 4.0±0.39 μm. Given its wide field of view, high resolution, and rapid slicing capabilities, LS-OS-SIM is anticipated to be extensively applied in the three dimensional imaging of both industrial microdevices and biological samples.

Compared with the traditional nickel-based materials and rare earth-based materials, the magnetostrictive coefficient is small, the eddy current loss is large, and the preparation cost is high. Utilizing cobalt ferrite as the driving core of magnetostrictive ultrasonic transducer can achieve stable service and high-power output at elevated frequencies. In this paper, a longitudinal magnetostrictive ultrasonic transducer with resonant frequency of 30 kHz is designed based on the magnetostrictive characteristics of cobalt ferrite and the finite element simulation is used to analyze the dynamics and magnetic field of the transducer. The output characteristics of the cobalt ferrite magnetostrictive ultrasonic transducer are tested by impedance matching. The resonant frequency, output amplitude and working temperature of the prototype are tested. The results show that the magnetic field intensity is 0.1~0.25 T (301~653 Oe), which can meet the driving requirement of cobalt ferrite in the best driving range(250~750 Oe).When the driving voltage is 12 V and the signal gain is 20 dB, the actual resonant frequency of the transducer is 29.8 kHz, the output amplitude is 3.75 μm, and the stable working temperature is 60 ℃. The experimental results verify the reliability and application potential of cobalt ferrite in the field of magnetostrictive transducers.

Taking all cities in Hunan province as the research object, the theoretical analysis framework of the integration level of cultural tourism industry was defined and the evaluation index system of the integration level of cultural tourism industry was constructed. The coupling coordination degree model, temporal evolution characteristics and spatial pattern characteristics visualization methods were used to measure the spatio-temporal evolution trend of the level of the cultural tourism industry, and the PVAR model was used to analyze the dynamic coupling relationship among the subsystems of the integration level of cultural tourism industry. The geo-detector model was used to explore the dominant factors affecting the integration level of cultural tourism industry and to detect the interaction effects among the factors. The main conclusions are as follows: 1) In 2009-2021, the integration level of culture tourism industry of cities in Hunan province can be divided into three types, the overall level shows an upward trend, but the differences within the region are also widened, and the expansion trend is significant. 2) During 2009-2021, the overall coupling coordination level of cities in Hunan province is still mainly in dysfunction class, with Changsha demonstrates a high degree of coordination class, Yueyang, Zhuzhou, Hengyang and Changde exhibit basic coordination class, and the remaining 9 cities are still in basic dysfunction class. 3) There exists an obvious dynamic coupling relationship among the subsystems of the industrial integration foundation, industrial integration support, and industrial integration volume in each city during the study period. 4) The 7 driving factors led by fiscal policy are the dominant factors influencing the integration level of cultural tourism industry of cities in Hunan province, and the factor interaction effects of fiscal policy, science and technology innovation and consumption capacity are strong.

A high-resolution reconstruction method of split-focal plane mosaic images is proposed due to the low spatial resolution of split-focal plane multispectral cameras. The method realizes the acquisition of a demosaicking reflectance image, which shoot a diffuse reflective white plate to calibrate a division-of-focal-plane (DoF) mosaic sensor. The reflectance image and the DoF mosaic image are substituted into the fusion model of multispectral and hyperspectral images for high-resolution reconstruction. The reconstruction results of two reconstruction methods based on superpixel segmentation operators (least squares-multispectral data fusion method and regularized minimum rank-multispectral data fusion method) are analyzed, and the root mean square error of the two methods is about 0.12. The results show that the demosaicking method achieves better results in the migration of the high-resolution fusion model for spectral images, and the DoF mosaic images are highly restored in both the spatial and the spectral information.

Multispectral ghost imaging (MGI) based on code-multiplexing is a novel imaging technology that significantly enhances spectral imaging performance through cleverly designed encoding and decoding strategies, characterized by simplification and intelligence. This paper starts from the fundamental principles and key technologies of the MGI and thoroughly explores various coding multiplexing strategies and their reconstruction algorithms, primarily including coding multiplexing strategies based on random patterns, Hadamard patterns, and Fourier patterns, as well as corresponding reconstruction algorithms such as compressed sensing, Fourier inverse transform, and deep learning. These approaches exhibit unique characteristics in extracting the spatial structure and spectral properties of the target scene, showing different imaging advantages and applicable scenarios. However, the technique still faces challenges such as low projection efficiency and high computational complexity. In the future, it is expected to further improve the imaging performance of the MGI system through more efficient image reconstruction algorithms, intelligent technologies and advanced optics to meet the high-quality imaging needs of complex scenes.

Multi-view imaging technology obtains rich scene information by capturing images from different angles, which can provide key visual data for fields such as autonomous driving, intelligent manufacturing and robot navigation. This paper first summarizes the components of common multi-view imaging systems, their working principles and implementation methods, and provides an in-depth analysis of the significant advantages of different multi-view imaging systems and their limitations from the key dimensions of dynamic imaging adaptability, imaging accuracy, cost-effectiveness, and comprehensive system performance. Then for the visual image enhancement technology, the application and effect in improving the quality of multi-view imaging are discussed by combining traditional image processing methods and deep learning techniques. Finally, the current status and future trends of the development of multi-view imaging technology are considered, and forward-looking predictions are made in the development directions of hardware innovation, algorithm optimisation and multi-modal data fusion.

As a driving force for new industries, the digital economy has emerged as a pivotal force in enhancing the efficiency of the tourism sector, critically contributing to the attainment of high-quality development in China’s tourism industry.Utilizing provincial panel data spanning from 2011 to 2020, this study employs the entropy weight TOPSIS and super efficiency SBM models to gauge the level of digital economy development and tourism industry efficiency. Furthermore, a comprehensive analysis is conducted using the two-way fixed effects model, adjustment effect model, and spatial Dubin model to explore the influence mechanism and effects of digital economy development on tourism industry efficiency. Findings reveal that digital economy development significantly boosts tourism industry efficiency, a conclusion upheld even after rigorous robustness tests. Analysis of regulatory mechanisms indicates that tourism intensity and the structure of the tourism industry positively moderate the relationship between digital economy development and enhanced tourism industry efficiency. Results from the spatial Dubin model demonstrate that the digital economy development within a province notably enhances its own tourism industry efficiency while exhibiting negligible spatial spillover effects on neighboring provinces’ tourism industry efficiency. Heterogeneity analysis further reveals that, compared with the eastern region, digital economy development in the central and western regions exerts a more pronounced influence on tourism industry efficiency.

The flood event has always been endangered high-quality development as for human society. Reconstructing and analyzing the impact-response process of the past flood events would be helpful to realize the mechanism and process of extreme climate events to human response, which could improve the adaptive capacity for natural disasters among human beings. In this study, archives and other historical documents were collected to reconstruct the spatial and temporal process of the flood event in 1811 Aksu. Furthermore, 26 points and 73 lines were extracted from above documents to reconstruct the impact-response process of this flood event, and social network analysis was applied to analyse the patterns of human interactions with climate change in the arid region. The main conclusions are as follows: 1) The flood event in 1811 Aksu was happened in the background of climate warming and humidification, and the interaction within the natural system among constant rainfall, ice and snow melt water, steep terrain exacerbated the impact of this flood event. 2) The main high-risk areas centralized in the Aksu city and adjacent villages, farmland. It was partly because the lower terrain and intense populations, various commodities near the city were prone to suffered by flood and waterlogging. 3) The flood affected the supporting system and population subsystem directly and indirectly affected the population subsystem, economic subsystem and social subsystem. In this process, there were 4 points (urban waterlogging, food production damaged, infrastructure damaged, famine) represented the main impacts of this flood event. 4) The human responses changed over time influenced by different disaster phases, including adjustment and adaption, which jointly effected to avoid the occurrence of post-disaster malignant events. And people constructed flood control engineering to prevent the occurrence of flood events that maintained the security and stability in northwestern frontier society during the middle and late Qing dynasty.

In the early stages of drug discovery, deep generative models are emerging as crucial tools for molecular design. The simplified molecular input line entry system(SMILES) serves as a standard chemical representation widely used for model training and generation. However, due to the non-uniqueness and non-directionality of linear representations of molecular ring systems, existing unidirectional encoders face limitations in capturing the global semantic structure of samples and generating valid molecular rings. Therefore, a method called Chemical RWKV BERT (ChemRB) is proposed, which aims to deeply extract bidirectional information from a large amount of unlabeled data. To achieve this, two pre-training tasks are innovatively designed: ring-level feature prediction and global-span closure prediction. These pre-training tasks not only provide the model with rich contextual information but also further enhance its in-depth understanding of the structural properties of complex molecules. Experimental results show that the ChemRB model not only achieves significant performance improvements but also reaches optimal baseline performance on new molecular/sample evaluation metrics. This excellent performance fully validates the effectiveness of ChemRB in accurately capturing the inherent structural information of molecules, providing a solid empirical foundation for its application in related fields. Finally, through testing and application on EGFR inhibitors, the practical utility and broad application prospects of the ChemRB model are further validated.

When mobile robots perform path planning, the traditional or classical ant colony algorithm often encounters problems such as fewer movement directions, smaller fields of view, non-optimal paths, and unsmooth paths. Aiming at the inherent shortcomings of the ant colony algorithm mentioned above, a parallel bidirectional 24 neighborhoods 16 directions ant colony algorithm is proposed. First, the 24 neighborhoods 16 directions path search method can expand the field of view of path search. Second, combining the 24 neighborhoods 16 directions path search method with the bidirectional alternating search strategy can better reach the endpoint and enhance the global search ability. Subsequently, the heuristic function includes starting point, current point, candidate node, and endpoint, as well as adaptive factors. At the same time, an improved transition probability formula is introduced to enhance the guidance of path search. Then, the crossing strategy is introduced to avoid getting stuck in local optima. Finally, the path node transfer strategy is adopted to smooth the path, resulting in fewer inflection points and the shortest path. On grid maps with different complexity, the improved ant colony algorithm proposed in this paper was compared with the traditional ant colony algorithm and other improved ant colony algorithms through simulation experiments. The simulation results proved that the algorithm proposed in this paper is feasible and effective.

The data from the fifth to seventh national censuses was used to explore a grid based method for agricultural population based on land use types, 1.5 km of agricultural population grid density data in Shaanxi province was obtained to reveal the current characteristics and spatiotemporal evolution process of agricultural population distribution. Research found that in 2020, the agricultural population in Shaanxi province was densely distributed in the Guanzhong Plain and Hanzhong Basin, while the Northern Shaanxi Plateau and Qinba Mountains were relatively sparse. At the same time, the agricultural population is relative densely distributed in river valleys, while the density is extremely low in urban core areas, difficult to use land, and ecological protection and water conservation areas. From 2000 to 2020, the overall agricultural population density in Shaanxi province has generally decreased, but there are differences in the decline rates among different regions, leading to increased spatial heterogeneity. The reason may lie in the different changes in land use types and the slowdown in agricultural population size. The accuracy verification found that the agricultural population grid data has good consistency with the census data, and the fitting accuracy of the agricultural population grid dataset is relatively high. The research has alleviated the problem of plasticity area units, fully captured the detailed information of agricultural population distribution, and has policy reference value for accelerating the shaping of reasonably distributed modern agricultural human resources and achieving comprehensive rural revitalization.

Aiming at the disadvantage that it is hard to obtain multiple Pareto sets in solving multimodal multi-objective optimization problems, a decomposition-based differential evolution algorithm is presented. In the proposed algorithm, multiple individuals that are assigned to the same weight vector form a subpopulation for finding multiple different Pareto sets. Then, an environmental selection method is designed to locate multiple different Pareto optimal solutions in the subpopulation. Finally, two differential evolution strategies are utilized to generate the offspring. The simulation results of the IEEE CEC 2019 benchmark test suite show that the proposed algorithm has good distribution ability in the decision space and can find more Pareto optimal solutions.

The visible optical polarization simulation data set in the scattering medium is relatively scarce, which limits the polarization imaging algorithm in the development and verification for the scattering medium. By considering the effects of scattering degeneration and light in the scattered environment on polarizing imaging, a polarizing image simulation method in the scattered medium is proposed. First, the physical render is used to simulate the propagation path of light and obtain the non-degraded polarization images through the polarized filter. After that, the real scattering environment data set (underwater and foggy environment) was used to determine the scattering degradation parameters in different environments. These parameters are then integrated with the scattering degradation model to generate simulated scattering polarization images. The simulation results closely match the image in real scattering environments. The DoLP(degree of linear polarization) images and AoP (angle of polarization) images reflect the phenomenon of depolarization in real scattering environments. Simulating polarization images can be used to analyze the polarization properties of the object and background in the scattered medium, and can provide a solid foundation for improving various polarized scattering algorithms.

In order to achieve non-destructive and rapid inspection of tissue microarrays, a fully automatic Mueller matrix microscopic imaging system was constructed by using the polarization imaging principle of the Mueller matrix. By integrating polarization optical components and imaging equipment with automatic control, this system can perform automatic imaging. The collected polarized image data is processed by using the Mueller matrix polar decomposition and Mueller matrix transformation algorithms to extract the key polarization characteristic parameters describing the sample. The results showed that in the experiment of distinguishing cervical cancer tissue samples from normal cervical tissue samples, the equivalent waveplate fast-axis azimuth θ and linear phase retardance δ are effective polarization parameters. By using statistical analysis, gray level co-occurrence matrix analysis and Tamura image processing methods to process the polarization parameter images, clear indications for diagnosing cervical cancer can be obtained. The fully automatic Mueller matrix microscopic imaging system can achieve rapid diagnosis of cancer by integrating multiple samples without frequently changing slices or switching to a high power objective.

Plant secondary metabolites are a class of biologically active organic compounds synthesized through secondary metabolic pathways during plant growth. These compounds are not essential for plant development but play crucial roles in many aspects including plant survival, environmental adaptation and stress resistance. The WRKY transcription factor family, named for its characteristic conserved WRKYGQK domain, is one of the largest and most functionally diverse transcription factor families in higher plants. These factors are widely distributed across the plant kingdom and regulate the synthesis of secondary metabolites, thus participating in various physiological processes including plant growth, environmental responses, and chemical defense. This review discusses the structural features and classification of WRKY transcription factors, with a focus on their role in regulating the synthesis of plant secondary metabolites such as terpenes, phenolics and alkaloids. Future research should delve deeper into the mechanisms by which WRKY transcription factors function within the plant secondary metabolism regulatory network and their potential applications in enhancing plant stress resistance and improving the quality of secondary metabolites. Such studies will provide theoretical and practical insights for the advancement of agricultural biotechnology and the functional improvement of plants.

Rotational ultrasonic effects were integrated with grinding and electrolysis effectively to achieve precision and high-efficiency processing of difficult-to-machine materials and complex shaped parts. A rotary ultrasonic composite electrolytic grinding system was designed and developed. Experimental schemes were designed to conduct mechanical grinding, ultrasonic grinding and rotary ultrasonic combined electrolytic grinding experiments on aluminum and SiCp/Al particle-reinforced ceramic materials. The forming mechanism of rotary ultrasonic combined electrolytic grinding is analyzed, and the influence mechanism of different parameters on machining precision, efficiency and surface quality is studied. Based on its mechanism, the parameters of rotary ultrasonic combined electrochemical grinding system were optimized. The results show that rotary ultrasonic combined electrochemical grinding can effectively reduce the cutting force and heat, ensure the removal of electrode materials and electrolyte circulation renewal in time, and enhance and stabilize the machining effect.When the voltage is 3 V, the precision of machining an aero-engine blade can reach 0.01 mm and the surface roughness can be less than 0.80 μm.The method of rotary ultrasonic combined electrochemical grinding has obvious technical advantages.

Compound erosion by multiple forces is the most significant feature of soil erosion on slope cropland in rolling hill regions with black soil in northeastern China. Compound erosion by wind and water is one of the important erosion types to aggravate the effects of soil erosion on slope cropland. In this study, on the experimental slope cropland in rolling hill regions, field observations of meteorological conditions, wind erosion and water erosion, and statistical methods were used to reveal the intra-annual alteration of erosion forces on compound erosion by wind and water and to analyze the intra-annual temporal succession characteristics of wind erosion and water erosion.Based on the threshold of wind speed for sand saltation (5 m/s) and erosive rainfall amount (12 mm/d) in rolling hill regions with black soil in northeastern China, the criteria for dividing the wind-dominated force and rainfall-dominated force of complex erosion on the experimental slope cropland was determined. The alteration of erosion forces of compound erosion by wind and water on the experimental slope cropland was divided into three stages: wind-dominated period Ⅰ(WDⅠ: Mid-March to Mid-July), rainfall-dominated period (RD: Mid July to Early October), wind-dominated period Ⅱ(WDⅡ: Early October to Mid-November). Surface creep erosion by wind accounted for about 92.95% of the total erosion amount by wind on the experimental slope cropland. Wind erosion rate in wind-dominated period was calculated at 432.9 g/m² in WDⅠ and 40.95 g/m² in WDⅡ. The rainfall erosion rate in rainfall-dominated period was estimated at 485.15 g/m².The wind erosion and water erosion on the experimental slope cropland showed temporal succession characteristics with the intra-annual alternations of erosion forces, following the order of water erosion rate in RD > wind erosion rate in WDⅠ>wind erosion rate in WDⅡ.

Most Transformer-based object tracking models have limited extraction of target's local spatial feature information and insufficient utilization of temporal features, significantly affecting the performance of object tracking models in handling complex scenarios such as target occlusion, deformation, or scale changes. Therefore, a visual object tracking method with spatial-temporal feature enhancement and perception (STFEP) are proposed in this paper. On one hand, this method uses Transformer for the extraction and fusion of search region and temporal context features to obtain global feature information. By designing a local convolutional neural network, it extracts the target's local feature information and associates it with the target's global feature information, further enhancing the target's feature representation. On the other hand, a spatial-temporal feature perception mechanism is proposed to analyze the reliability and necessity of feature information at different moments, constructing dynamic templates to perceive richer spatial-temporal information, enabling the model to adapt to complex changes in targets and scenes. Experimental results on multiple datasets such as TrackingNet, GOT-10k, LaSOT and UAV123 show that the proposed method can track the target accurately and robustly, and the optimal results are obtained on GOT-10k dataset. AO, SR_0.5 and SR_0.75 were 73.7%, 83.8% and 70.6%, respectively.

Infrared polarization imaging detection technology introduces polarization information on the basis of traditional intensity information, which can effectively improve target detection and recognition capabilities under specific conditions. It has advantages such as high signal-to-noise ratio, anti camouflage, and anti-interference, and has broad application potential and good development prospects in target reconnaissance, detection, and strike fields. This article first introduces the theory of surface polarization of objects and related phenomena, especially the discovery of high-temperature polarization phenomenon of targets, which expands new fields for polarization detection applications.Based on the theory of target polarization, the research results of target polarization characteristics under different environments and application backgrounds were analyzed. Secondly, the development of infrared polarization detectors and the latest progress in target infrared polarization imaging detection in recent years were reviewed. The new requirements for infrared polarization detection technology in the current complex battlefield environment were summarized, including breaking through the bottleneck of real-time high-precision imaging technology, strengthening cloud and fog penetration, and anti occlusion interference performance.Finally, based on the summary of the target polarization mechanism, target polarization characteristics, and the development trend of polarization detector technology, it is proposed to deepen the theoretical research on target surface polarization, promote the development of high-precision infrared polarization detector preparation technology, explore multi-dimensional information fusion processing, and further look forward to the future development and application prospects of infrared polarization detection technology in military fields such as complex target recognition, anti stealth operations, and battlefield monitoring.

To establish a tourism-friendly bay area, it is essential to assess the cross-sea accessibility of the Pearl River Estuary, thereby effectively addressing the connectivity issues between the eastern and western shores of the bay area. Utilizing the cost-weighted grid method, this study examines the impact of newly built cross-sea channels on the accessibility of tourist attractions in the Greater Bay Area (GBA) from four aspects: changes in isochrone area, daily accessibility, weighted average travel time, and tourism economic linkage intensity. Taking the Hong Kong-Zhuhai-Macao Bridge and the Shenzhen-Zhongshan Channel (hereinafter referred to as “B&C”) as examples, the study analyzes their influence. The conclusions are as follows: 1) Within a certain spatial scope, there is a marginal diminishing effect on the improvement of tourist attraction accessibility, and the enhancement effect declines with distance from the starting and ending points of the cross-sea tunnels. The main contradiction in improving GBA accessibility has shifted from absolute value enhancement to controlling the expansion of regional disparities and structural imbalances. 2) The impact of B&C on tourism accessibility is concentrated mainly on the eastern and western shores of the Pearl River Estuary, with insignificant effects on the central region. Specifically, the HKZMB primarily enhances the area of the 2-hour traffic circle in the GBA, while the Shenzhen-Zhongshan Channel mainly expands the 1-hour traffic circle. 3) In terms of attraction accessibility, B&C significantly increase the number of attractions reachable within 1 hour on both shores of the Pearl River, enabling the areas around the Pearl River Estuary in the GBA to form a relatively close tourism circle. The overall attraction accessibility is compressed from 1.20 hours to 1.12 hours, with Zhaoqing, Jiangmen, and Huizhou representing weaknesses. 4) Measuring the changes in the tourism economic landscape of the GBA based on tourism economic linkage intensity reveals that B&C have not altered the three core clusters of Guangzhou-Foshan, Shenzhen-Hong Kong, and Zhuhai-Macao, but rather further consolidated their agglomeration trends.

Multi-functional active peptide is a protein-derived compound that can act on multiple targets and deliver a variety of physiological effects, and has significant therapeutic effects on a variety of diseases. The existing multi-functional active peptide prediction model fails to fully consider the correlation between amino acids in the feature representation stage, which reduces the feature representation ability of the model, and the existing method adopts the strategy of converting the multi-label classification problem into multiple binary classification prediction problems, which leads to the inability of the model to consider the dependence between multiple functions of the active peptide in the prediction stage, which reduces the prediction accuracy of the model for multi-functional active peptides. In order to solve the above problems, a multi-functional active peptide prediction model based on label dependence is proposed, TCLD, which extracts the correlation between amino acids in the active peptide sequence through the Transformer encoder, and uses the ZLPR loss function to capture the dependence between multiple functions, which is used to improve the performance of the multi-functional active peptide prediction model. The experimental results show that the prediction performance of TCLD is better than that of the existing multi-functional active peptide prediction methods, which is helpful for researchers to quickly screen out multi-functional active peptide candidates with potential therapeutic value, thereby accelerating the research and development process of new drugs.

Janus-Helmholtz (JH) transducers and Janus-Hammer Bell (JHB) transducers have similar appearance and structure, but their vibration modes are different. This article firstly analyzes the structural parameters corresponding to different vibration modes in JH transducer and JHB transducer. Under the conditions determined by other parameters, through theoretical calculation, it is concluded that the relationship between the axial length L of the ring shell in the JH transducer and the inner radius a of the ring shell should satisfy 2<L/a<5.7. The value range corresponding JHB transducer is 0.1<L/a<0.3. Then the finite element special boundary condition setting method is used to analyze the vibration modes corresponding to the resonance peaks of the JH transducer and the JHB transducer. It is concluded that the lower-frequency vibration mode in the JH transducer is the Janus longitudinal vibration, and the high-frequency vibration mode is the cavity resonance mode.The lower-frequency vibration mode in the JHB transducer is the Janus longitudinal resonance, and the higher-frequency vibration mode is the ring breathing mode. Finally, the radial and axial transmitting voltage responses of the JH transducer and the JHB transducer are compared to analyze work performance under different working modes, and it is concluded that the JH transducer can achieve a wider operating bandwidth and a higher transmitting voltage response. The working bandwidth of JHB transducer is narrow, but the transmitting voltage response in the radial direction has small fluctuations, and the transmitting voltage response in the axial direction has large fluctuations.

Polarization spectral imaging technology provides an effective solution to the imaging problems in complex scenes by combining intensity, polarization, and spectral information. Division-of-focal-plane polarization spectral imaging technology has become a significant development direction in this field due to its high compactness and strong real-time performance. This paper first reviews the development history of polarization spectral imaging technology and systematically compares the advantages and disadvantages of related technologies. Then, it focuses on the research progress of the division-of-focal-plane polarization spectral imaging system, providing a detailed overview of polarization spectral splitting elements, polarization spectral demosaicking algorithms, and existing polarization spectral image databases. It also systematically summarizes the core advantages of this technology, including high compactness, strong real-time performance, and low power consumption. Finally, it summarizes the applications of polarization spectral imaging technology in military reconnaissance, space exploration, medical diagnosis, and remote sensing detection. The analysis indicates that this technology has broad prospects in fields such as target detection, environmental monitoring, and medical diagnosis, but it still faces challenges such as low spatial resolution and insufficient accuracy in reconstructing polarization information currently. Based on this, future research should focus on optimizing the design and fabrication of polarization spectral splitting elements, improving demosaicking algorithms for high-quality image restoration, and further expanding its application capabilities in dynamic scenes and complex environments.

The polarization parameter images in polarization dehazing algorithms are highly susceptible to noise interference, and the brightness of the dehazed images is unstable. This leads to the algorithm’s inability to adapt universally to different lighting conditions. To address this issue, a new polarization dehazing algorithm is proposed, which introduces a novel polarization parameter image based on low-rank approximation to reduce noise interference. Additionally, a multi-exposure fusion method is employed to compress the image’s dynamic range. Comparative experiments demonstrate that this method effectively removes haze from images, exhibiting strong robustness and significantly improving the overall quality of haze-affected images captured in various environments. In terms of standard deviation, image information entropy, NIQE and PIQE, the indicaters of the proposed method are improved by 22.99%,4.06%,17.42%,32.89% and 33.33%,2.80%,12.31%,76.14% respectively compared with the method proposed by Schechner and dark channel method.

Dietary biomarkers offer an objective means of assessing contemporary food exposures, serving as an alternative or supplementary tool to self-reported food intake. They are also utilized to explore the functional characteristics of dietary factors. The precise measurement of dietary biomarkers promotes our understandings of the relationship between diet and risk of developing metabolic diseases, helping to achieve personalized and precise nutrition. Metabolomics is a key method for screening dietary biomarkers. This article reviews the research results on dietary biomarkers related to the intake of grains, fruits and vegetables, meat, fish, dairy products, and nuts based on large-scale population cohort studies and randomized controlled dietary intervention studies from 2013 to 2023. It also summarizes the relationship between specific dietary biomarkers and the risk of 35 dietary nutrition related metabolic diseases, including obesity, cardiovascular and cerebrovascular diseases, cancer, and Alzheimer’s disease. The current status and challenges of discovering dietary biomarkers based on metabolomics are discussed, in particular on the issues related with research design, complex food composition, inter-individual differences and metabolomics data processing techniques. This review emphasizes the enormous potentials and research prospects of using dietary biomarkers to uncover the relationship between diet and health.

Lead-free metal halide biperovskite is composed of non-toxic elements, stable in air and has a long carrier lifetime. The physical properties of bismuth based Cs₂MBiX₆(M=Cu, Ag, Au, X=Cl, Br, I) double perovskite materials with excellent photovoltaic properties were calculated theoretically.In order to analyze the effects of different lead-free metal halides on battery performance, first-principles calculations were performed to systematically investigate the crystal structures,electronic structures and optical properties of four materials Cs₂AgBiI₆, Cs₂AuBiCl₆, Cs₂CuBiBr₆, and Cs₂AgBiBr₆.Finally, the absorption rate, carrier collection efficiency, external quantum efficiency, short-circuit current density, open circuit voltage and volt-ampere characteristics for the layered architecture consists of FTO/c-TiO₂/Cs₂MBiX₆/spiro-OMeTAD/Au structure perovskite solar cell are analyzed by performing equivalent optical admittance method. The results show that:when the thickness of the absorption layer is 0.6 μm.The short-circuit current densities of perovskite solar cells prepared with Cs₃AgBiI₆, Cs₂AuBiCl₆, Cs₂CuBiBr₆ and Cs₂AgBiBr₆ are 27.6, 26.0, 22.3 and 10.9 mA/cm², respectively, corresponding to open circuit voltages of 0.83, 0.87, 1.08 and 1.1 V. The photoelectric conversion efficiency of the device is 19.3%, 16.6%, 21.3% and 10.9%, respectively.It is found that 4 kinds of materials have high thermodynamic stability, suitable band gap and high absorption coefficient of ~10⁵cm^-1 in the visible light range, and the cell with optimized device structure also has considerable photoelectric conversion efficiency.

Rural tourism is an important engine for achieving the strategy of rural revitalization. Exploring the spatio-temporal evolution laws of rural tourism destinations is of great significance for promoting high-quality development of rural tourism. Based on the theory of “production-living-ecological” function, 303 provincial-level key rural tourism villages in Shaanxi province are taken as research objects to study the temporal-spatial distribution characteristics and influence mechanism of rural tourism destinations in Shaanxi province by using the methods of GIS spatial analysis and geographical detector. The conclusions are as follows: 1)The spatial distribution of all rural tourism destinations in Shaanxi province is characterized by gathering in mid to low altitude terrain, near the main transportation line, along rivers and lakes, showing a developing trend of polarization and sheet diffusion. Rural tourism destinations with dominant function production have the greatest spatial dependence on transportation and hydrological conditions.2)There are certain differences in the spatial distribution and evolution of rural tourism destinations dominated by different functions. Rural tourism destinations primarily driven by production functions exhibit the highest degree of agglomeration, displaying a contiguous expansion trend, and are predominantly situated in areas with flat terrain and dense river networks.The distribution of rural tourism destinations driven by life functions is largely consistent with the overall distribution, and there is a noticeable increase in the agglomeration of the high-density core areas.Rural tourism destinations driven by ecological functions exhibit a clear trend of outward expansion. These destinations are predominantly located in the border regions between cities, forming a point-to-axis distribution pattern. Additionally, a high-density distribution area has emerged at the southern foothills of the Qinling Mountains.3)There are differences in the influencing factors of spatial differentiation of rural tourism destinations dominated by different functions. Relevant policies exert the strongest influence on the spatial distribution of rural tourism destinations dominated by production functions. Rural tourism destinations dominated by life functions have a high spatial dependence on population distribution and surrounding scenic areas, while rural tourism destinations dominated by ecological functions are significantly affected by climate conditions and air quality.4)Natural ecological environment determines the basic spatial distribution pattern of rural tourism destinations, economic development level is an important driving force for the formation and development of rural tourism destinations, social and political conditions are guiding factors. Resource endowment foundation plays a prominent driving role in the spatial distribution differentiation of rural tourism destinations across various functional types. Furthermore, the interactive enhancement effect with economic development level surpasses that with social and political conditions.

Mixed attribute data is one of the most common types of datasets, and clustering algorithms tailored for this type of data are a research hotspot in clustering analysis. Due to the advantages of spectral clustering algorithms in handling clustering problems of arbitrary-shaped data and converging to global optimal solutions, an improved spectral clustering algorithm(improved Jaccard and Mahalanobis-spectral clustering, IJM-SC) from the perspective of similarity measurement formulas is proposed.Based on the ideas of Jaccard distance and Mahalanobis distance, a similarity measurement suitable for mixed attribute data is designed, and its application in spectral clustering of mixed attribute data is explored. The developed algorithm is applied to cluster three mixed attribute datasets including the UCI heart disease dataset, demonstrating its superiority in clustering mixed attribute data. By comparing the performance metrics with existing algorithms, the results demonstrate that the proposed algorithm achieves better clustering of mixed attribute data.

In order to explore the water extraction process and quality standard of compound Juncao Ganoderma lucidum granules, liquid extract yield and liquid extract polysaccharide content were taken as indexes, the effects of extraction temperature, extraction time, ratio of material to liquid, extraction times and soaking time on extraction yield and polysaccharide content were investigated by single factor and response surface experiments. The particle size, moisture content, weight loss on drying, loading difference, solubility, microbial limit and heavy metal content of granules were detected according to the 2015 edition of Pharmacopoeia of the People’s Republic of China and the national standard. The raw materials of Juncao Ganoderma lucidum, Perilla frutescens and selenium-enriched tea in compound Juncao Ganoderma lucidum granules were identified by thin layer chromatography, and the contents of monosaccharides and tea polyphenols in compound Juncao Ganoderma lucidum granules were determined through high performance liquid chromatography. The results showed that the optimal extraction conditions were as follows: extraction temperature 80 ℃, extraction time 2.5 h, ratio of material to liquid(g/mL) 1∶25, two times of extraction, and immersion time 6 h. The particle size, water content, solubility, weight loss on drying and loading difference of the granules reached the requirements of the 2015 edition of Pharmacopoeia of the People’s Republic of China.The microbial limit and heavy metal content of the granules met the requirements of the national standard, and the average selenium content was 0.058 3 mg/kg. The TLC bands of Juncao Ganoderma lucidum, Perilla frutescens and selenium-enriched tea were clear and well separated, with no interference from the negative control. The contents of mannose, rhamnose, glucose, galactose, arabinose, epigallocatechin gallate, epicatechin gallate, catechin, gallic acid in compound Juncao Ganoderma lucidum granules were 1.310, 1.166, 8.708, 1.265, 0.824, 5.79, 2.58, 1.72 and 1.67 mg/g, respectively. The optimized water extraction process of the compound Juncao Ganoderma lucidum granules is stable and reliable, all the indexes of the granules meet the requirements, and the established quality standard for the water extraction process of compound Juncao Ganoderma lucidum granules is feasible.

In the process of interference fit, in order to make the contact surface fit more closely and reduce the wear between the contact surfaces, an ultrasonic vibration-assisted press fit structure is designed. The dimensions of the horn are determined based on the matching of the output end of the horn with the press fit component, and the ultrasonic vibration frequency and power of the device are changed by altering the type and quantity of the vibrator. The results indicate that after the introduction of ultrasonic vibration, the press fit stress on the inner surface of the hub decreases by 20%, and the stress on the outer surface of the bearing decreases by 17%. With the increase of ultrasonic frequency and power, the press-fit force decreases by 10% to 20%. The greater the ultrasonic vibration power and the higher the frequency, the smaller the press fit force required for interference fit. The introduction of ultrasonic vibration in the press fit process effectively improves the connection performance of the interference fit and reduces the interface loss of the connection surface.

To improve the segmentation accuracy of cataract surgical instruments, an EE-DANet was constructed in this paper. EE-DANet adopts a double-branched structure: The edge branch is responsible for extracting edge features, using augmented spatial attention (ASA) to address the issue of edge information loss in traditional image segmentation; The main branch adopts a U-shaped structure, and the multi-scale feature fusion module is used in the feature extraction stage to solve the problem of multiple types of surgical instrument images and large scale transformation; in the decoder section, a strip coordinate attention (SCA) is used to globally model the image, capturing the direction and position information of the surgical instrument, and address the problem of mirror reflection of the surgical instrument. The IoU and Dice reached 75.5% and 83.9% on the Dataset_instrument dataset, respectively. The proposed model effectively improves the segmentation performance of cataract surgery instruments, and has a positive significance for the clinical diagnosis of cataracts.

It is a typical application of ultrasonic machining to superimpose ultrasonic vibration on micro tools for precision machining of small structural parts. In ultrasonic-assisted machining, the ultrasonic vibration of the tool is an important parameter that affects the machining quality. However, due to their micro diameter, measuring directly or indirectly the ultrasonic vibration of micro-tools with complex geometry by using the exist method poses a current challenge. To solve the problem, an acoustic characterization method for the micro-tool’s ultrasonic vibration is proposed. The relationship between the tool ultrasonic vibration and the radiated sound pressure is studied theoretically. Additionally, the differences between the measured sound pressure under different measuring conditions and the ideal sound pressure radiated by the tool are analyzed, and the sound pressure distribution radiated by different micro-tools is measured. Then a sound pressure measurement strategy is proposed and the corresponding experiment platform is established to characterize the ultrasonic vibration of the micro-tool. By using the acoustical measurement strategy, the working frequency of micro twist drill with different diameters(0.1~0.9 mm) is identified and the relative amplitude of tool vibration is quantified. The method based on radiative sound pressure can be used to in situ characterize the ultrasonic vibration state of micro-tools at low cost and high adaptability, which supplies a fundamental tool for the ultrasonic-assisted micro-machining process study.

This study utilized eight kiwifruit cultivars as materials to investigate kiwifruit quality traits, including soluble solid content (SSC), soluble sugar content, organic acid content, vitamin C content and aroma components. These traits were analyzed in conjunction with resequencing data. The results indicated that the SSCs of the eight kiwifruit cultivars ranged from 10.90% to 19.28%, the soluble sugar contents ranged from 80.24 g/kg to 132.82 g/kg, and the organic acid contents ranged from 7.28 g/kg to 30.99 g/kg. Among these kiwifruits, the average single fruit weight of Actinidia chinensis was significantly higher than those of the other species. A. eriantha exhibited the highest vitamin C content, reaching up to 6 541.32 mg/kg.A. arguta had the highest sugar-acid ratio. The main aroma components of A. chinensis, A. deliciosa and A. rufa were esters, whereas the aroma components of A. arguta and A. eriantha were mainly alkenes. The hybrid cultivars ‘Jinyan’ and ‘Mantianhong’ also showed esters as their main aroma components. By comparing the quality parameters of the hybrid kiwifruit cultivars with those of their parental species, it was demonstrated that the quality parameters of ‘Jinyan’ and ‘Mantianhong’ were closer to those of A. chinensis.Genetic clustering analysis also indicated that the two hybrid cultivars were more closely related to their paternal species A. chinensis.

The traditional flower pollination algorithms tend to exhibit poor exploitation when facing some complex optimization problems. Aiming at this weakness of the traditional flower pollination algorithms, an adaptive guidance flower pollination algorithm (AGFPA) is proposed. In the proposed AGFPA, an adaptive guidance mechanism (AGM) is introduced, which combines the global best individual surrounding strategy and the global best approaching strategy.The introduced AGM adaptively utilizes the global best individual to guide the population evolution, enhancing the exploitation capabilities while preserving the population diversity as much as possible. Specifically, the global best individual surrounding strategy focuses on exploiting the neighborhoods around the global best individual. Meanwhile, the global best approaching strategy utilizes the global best individual to guide the search directions, enabling the algorithm to explore a wide unknown area. In addition, an adaptive parameter control strategy is presented in the proposed AGFPA. The two key parameters, global pollination transform probability and step size factor, are adjusted according to the needs of different evolution stages, maintaining a good balance between exploitation and exploration. To test the performance of AGFPA, 18 benchmark functions are utilized in the experiments, which are commonly used in the field of swarm intelligence. The effectiveness of the strategies is discussed. Moreover, AGFPA is compared with several existing flower pollination algorithms and particle swarm optimization algorithms. Additionally, AGFPA is also used to estimate the fermentation kinetic parameters in the biochemical engineering. The experimental results show that AGFPA can exhibit promising performance on the most unimodal, multimodal and complex functions. Moreover, AGFPA can yield excellent results in the biochemical engineering applications.

Calcineurin B-like proteins (CBLs) are involved in plant-specific Ca²⁺ signaling as calcium (Ca) sensors and are essential for plant response to abiotic stress. Peanut is an important oilseed crop, and salt stress seriously affects the growth and development of peanut and its seed quality.At present, the response of peanut CBL genes to salt stress is unknown. In this study, members of the CBL gene family were identified from the peanut genome-wide database using bioinformatics methods and were analyzed for physicochemical properties, gene structure, phylogenetic tree and colinearity. The study found that there are 89 CBL genes in the peanut genome distributed on 19 chromosomes. Analysis of the physicochemical properties of the proteins showed that peanut CBLs are mainly composed of acidic amino acids. Subcellular localization predictions revealed that most of the CBLs in peanut are localized in the nucleus. To investigate the response of peanut CBL genes to salt stress, the expression patterns of five CBL genes were analyzed under salt stress. After treating with 200 mmol/L NaCl for 3 h, the expression of AhCBL46.1 was significantly increased, and the expressions of AhCBL4, AhCBL33 and AhCBL66 were significantly increased after 12 h of NaCl treatment. However, AhCBL61.3 showed a negative response to salt stress, suggesting that the peanut CBL genes play an important role in salt stress regulation. This research lays the groundwork for further exploration into the functionality of AhCBL genes.