In recent years, there has been a significant development in unmanned platform technologies, specifically unmanned ground vehicles (UGVs) and unmanned aerial vehicles (UAVs). As a result, their application scenarios have expanded considerably. Unmanned platforms are considered integral components of the Internet of Things system. However, certain challenges arise when dealing with specialized tasks, such as navigating complex urban low-altitude terrain with multiple obstacles and limited communication capabilities. These challenges can greatly impact the efficiency of the system due to information isolation. To address this issue, a messenger drone mechanism is introduced in this paper, which utilizes air superiority to facilitate indirect communication between unmanned platforms. Additionally, a task sequence planning algorithm based on sampling transformation is designed. This algorithm efficiently assigns the drone to mobile UGVs by discretely sampling their paths and considering the UAV-UGV motion relationship. By transforming the problem into an asymmetric traveler problem, it allows for a fast solution. Finally, the effectiveness of the algorithm is verified through comparative analysis in different scenarios.

To accurately determine the leakage source location and strength during gas leakage accidents, this study compares the concentration obtained from the diffusion model with that measured by the sensor and proposes an improved gray wolf optimization algorithm for leakage source location. This algorithm introduces two improvement strategies. First, a nonlinear convergence factor is introduced to balance the global and local searches of the algorithm. Second, a reverse learning operation is performed on the three individuals with the worst fitness in the contemporary population. The results showed that the location results based on the improved gray wolf optimization algorithm exhibited high accuracy and stability, could quickly and accurately locate the leakage source, and provided data support for emergency disposal of accidents.

The structural effects of monetary policy and these effects impact on credit allocation are crucial facets of monetary policy research. This study initially estimates China’s monetary policy based on the GDP growth target and CPI target found in the Chinese Government’s Annual Work Report. Subsequently, it quantifies the extent of credit misallocation among Chinese firms using data from listed companies from the years 2013 to 2022. Finally, we empirically investigate the repercussions of expansionary monetary policy shocks on credit misallocation, focusing on micro-firms. Empirical findings reveal that expansionary monetary policy significantly exacerbates credit misallocation, particularly in underfinanced firms. Mechanistic analysis suggests that a preference for loan size in lending behavior constitutes a major factor contributing to credit misallocation among firms.

In recent years, several attempts have been made to quantitatively evaluate covert attention using microsaccades. However, several unclear aspects exist regarding the measurement method of microsaccades, and a unified analysis method does not exist. Therefore, the current status is such that the interpretation of the results is divided among the research groups. To address this problem, empirical studies on microsaccades must be accumulated and measured and evaluated using a unified method. Therefore, in this study, to accumulate empirical studies on microsaccades, an experiment was conducted to investigate the effect of the presence or absence of gazing at a fixation point on the interval of occurrence of microsaccades in a measurement task. The participants were 15 healthy young people, and we compared the following two types of measurement tasks. Task-I: The participants freely visually searched a white wall 1 m away for 120 s. Task-II: The participants gazed at a fixation point located 1 m ahead at eye level for 120 s. For the microsaccade detection algorithm, we adopted a method imitating the EK method proposed by Engbert and Kliegl in 2003, divided 120 s of time-series data and analyzed it every 2 s, and subsequently evaluated the time-series data for the entire 120 s by concatenating them. Consequently, the interval of occurrence of microsaccades during Task-II decreased by more than 1 Hz compared with that of Task-I (p<0.05). The study confirmed that the presence or absence of gazing at the fixation point during microsaccade measurement affected the interval of occurrence of microsaccades.

Despite the medicinal significance of traditional Thai herbs, their accurate identification on digital platforms is a challenge due to the vast diversity among species and the limited scope of existing digital databases. In response, this paper introduces the Thai traditional herb classifier that uniquely combines transfer learning, innovative data augmentation strategies, and the inclusion of noisy data to tackle this issue. Our novel contributions encompass the creation of a curated dataset spanning 20 distinct Thai herb categories, a robust deep learning architecture that intricately combines transfer learning with tailored data augmentation techniques, and the development of an Android application tailored for real-world herb recognition scenarios. Preliminary results of our method indicate its potential to revolutionize the way Thai herbs are digitally identified, holding promise for advancements in natural medicine and computer-assisted herb recognition.

At the present stage, the identification of dangerous behaviors in public places mostly relies on manual work, which is subjective and has low identification efficiency. This paper proposes an automatic identification method for dangerous behaviors in public places, which analyzes group behavior and speech emotion through deep learning network and then performs multimodal information fusion. Based on the fusion results, people can judge the emotional atmosphere of the crowd, make early warning, and alarm for possible dangerous behaviors. Experiments show that the algorithm adopted in this paper can accurately identify dangerous behaviors and has great application value.

Integrated energy microgrids and shared energy storage have significant benefits in improving the energy utilization of the system, which is gradually becoming the current research hotspot. And the uncertainty of new energy output also significantly affects the stable and economic operation of integrated energy microgrid. So how to establish a set of integrated energy microgrids optimization operation model considering photovoltaic (PV) output uncertainty and shared energy storage is an urgent problem to be solved nowadays. Firstly, this paper introduces the framework of an integrated energy system microgrid containing a shared energy storage operator (ESO), and analyzes the scheduling method of the upper tier operator within the system as well as the economic benefits at the lower tier user end. Secondly, to address the randomness of PV output, Monte Carlo method is used to generate the scenarios, and then the scenarios are cut down by using the fast antecedent elimination technique. Then, an optimal operation model is established for micro grid operator (MGO) and user aggregator (UA), respectively, and based on the master-slave game relationship, so that the MGO is the leader and the UA is the follower, a Stackelberg game model is proposed to consider the integrated demand response of electricity and heat between the MGO and UA in the context of the participation of ESO in the auxiliary service of the UA. Finally, the proposed model is brought into a typical residential building community for simulation verification, and the results show that the model proposed in this paper can effectively balance the interests of MGOs and UAs, and realize win-win benefits for UA and ESO.

Classifying customers according to their characteristics can effectively meet the genuine needs of different customer groups. It also helps enterprises formulate reasonable marketing strategies and obtain considerable profits. Currently, there are many ways to classify customers. However, the procedures involved are complicated and cannot comprehensively and objectively reflect customer characteristics. Therefore, a customer group classification model is designed based on the deep cross network (DCN). The DCN algorithm can automatically learn simple data features, achieving data clustering. For the defects in this model, the deep weighted k-means clustering network (DWKCN) customer group classification method is constructed, improving the DCN algorithm. From the results, the algorithm has a high accuracy of 99.5%. Therefore, the proposed DWKCN algorithm can realize the customer group’s precise division and the marketing plan design, providing the references for different types of customers to formulate personalized needs.

Basketball has rapidly developed in recent years. Analysis of various moves in basketball can provide technical references for professional players and assist referees in judging games. Traditional technology can no longer provide modern basketball players with theoretical support. Therefore, using intelligent methods to recognize human body postures in basketball was a relatively innovative approach. To be able to recognize the basketball sports posture of players more accurately, the experiment proposes a basketball stance recognition model based on enhanced graph convolutional networks (GCN), that is, the basketball stance recognition model based on enhanced GCN and spatial temporal graph convolutional network (ST-GCN) model. This model combines the respective advantages of the GCN and temporal convolutional network and can handle graph-structured data with time-series relationships. The ST-GCN can be further deduced by realizing the convolution operation of the graph structure and establishing a spatiotemporal graph convolution model for the posture sequence of a person’s body. A dataset of technical basketball actions is constructed to verify the effectiveness of the ST-GCN model. The final experimental findings indicated that the final recognition accuracy of the ST-GCN model for basketball postures was approximately 95.58%, whereas the final recognition accuracy of the long short term memory + multiview re-observation skeleton action recognition (LSTM+MV+AC) model was about 93.65%.

Currently, numerous high-precision models have been proposed for semantic segmentation, but the model parameters are large and the segmentation speed is slow. Real-time semantic segmentation for urban scenes necessitates a balance between accuracy, inference speed, and model size. In this paper, we present an efficient solution to this challenge, efficient asymmetric attention module net (EAAMNet) for the semantic segmentation of urban scenes, which adopts an asymmetric encoder–decoder structure. The encoder part of the network utilizes an efficient asymmetric attention module to form the network backbone. In the decoding part, we propose a lightweight multi-feature fusion decoder that can maintain good segmentation accuracy with a small number of parameters. Our extensive evaluations demonstrate that EAAMNet achieves a favorable equilibrium between segmentation efficiency, model parameters, and segmentation accuracy, rendering it highly suitable for real-time semantic segmentation in urban scenes. Remarkably, EAAMNet attains a 73.31% mIoU at 128 fps on Cityscapes and a 69.32% mIoU at 141 fps on CamVid without any pre-training. Compared to state-of-the-art models, our approach not only matches their model parameters but also enhances accuracy and increases speed.

Most of the existing trackers based on discriminative correlation filters use only one feature or a simple linear fusion of multiple features for object tracking, and most of them lack a mechanism to handle occlusions. This leads to poor tracking performance in rapidly changing and easily occluded scenarios, especially on unmanned aerial vehicle (UAV) platforms. To address this issue, this paper proposes an anti-occlusion visual tracking algorithm for UAVs with multi-feature adaptive fusion named multi-feature adaptive fusion and anti-occlusion tracker (MAFAOT). It introduces a novel approach for implementing an adaptive fusion of multiple features. This method transforms the multi-feature fusion problem into a maximization issue by designing a tracking quality evaluation index. It successfully achieves an adaptive fusion of gradient direction histogram and color histogram feature responses. MAFAOT also introduces an anti-occlusion update pool strategy, enabling the tracker to adapt dynamically to various complex scenarios, including occlusion and motion blur. The experimental results on the OTB100 and UAV123 datasets confirm the significant advantages of MAFAOT in terms of precision and success rate compared to other correlation filter-based algorithms. The proposed methods further enhance the expressiveness of the features and effectively avoid the problem of tracker contamination caused by occlusion. Furthermore, this paper applies the proposed methods to the kernelized correlation filters (KCF) algorithm. On the OTB100 dataset, the improved KCF algorithm shows an improvement of 10.94% in precision and 11.11% in success rate. On the UAV123 dataset, it shows an improvement of 14.53% in precision and 16.62% in success rate, further verifying the effectiveness and versatility of the proposed methods.

To solve the problem of poor localization accuracy and robustness of visual simultaneous localization and mapping (SLAM) systems in highly dynamic environments, this paper proposes a dynamic visual SLAM algorithm called FFD-SLAM that fuses the target detection network with the optical flow method. The algorithm considers ORB-SLAM2 as the basic framework, joins the semantic thread in parallel with its tracking thread, initially obtains the set of feature points through the real-time detection of dynamic objects in the environment through YOLOv5 in the semantic thread, then filters the set of feature points obtained in the semantic thread through the optical flow module, and finally utilizes the remaining static feature points for the matching calculation. Experiments showed that the proposed algorithm showed an improvement of approximately 97% in the localization accuracy compared with the ORB-SLAM2 algorithm in a highly dynamic environment, which effectively improves the localization accuracy and robustness of the system. The proposed algorithm also showed a higher real-time performance compared with some excellent dynamic SLAM algorithms.

This paper introduces a new lower-limb rehabilitation machine that meets the rehabilitation needs of hemiplegic patients. First, a left–right independent rotary pedal mechanism was selected to facilitate rehabilitation and adapt to the user’s physical condition. Then, a half model of the lower-limb rehabilitation machine is designed and manufactured with ergonomics in mind. As analytical tools, we combine non-negative matrix factorization and non-negative double singular value decomposition to calculate muscle synergy of the walking muscle surface electromyography (sEMG) signal, and use cosine similarity to evaluate the similarity between walking and pedaling activities. By comparing the results of the walking and pedaling experiments, the effectiveness of pedaling in gait rehabilitation is revealed. To further improve the similarity between walking and pedaling, double integration of the sEMG signal is introduced, and the relationship between load input and rotation angle is described for the first time using Fourier series. The results of the experiment confirmed that more than half of the 10 subjects performed pedaling exercises similar to walking using Fourier series loading compared to pedaling exercises with normal constant loading. This loading parameter may have the potential to improve rehabilitation efficiency for many subjects compared to the usual exercise.

As coal ore and other resources are continuously mined, a three-zone structure is formed underground consisting of a sagging zone, fault zone, and caving zone. The use of well-logging data to identify the three zones is important for production safety and environmental management. Owing to the scarcity of data that can reflect three zones in normal coal mining, conventional identification and prediction methods face challenges when extracting data features, incurring a degree of uncertainty within prediction results. Accordingly, the accurate identification of the three zones has become a critical objective in daily production. To address this issue, we developed a method called a method called backpropagation neural networks with Dempster–Shafer (DS) evidence theory. Initially, we preprocessed the training data and deployed two backpropagation neural networks (BPNNs) to predict the three zones according to two parameters. According to these prediction results, the local and global credibility of each prediction is calculated and used to obtain the basic probability assignment function required for the DS evidence theory. Finally, the DS evidence theory is used to fuse the two BPNNs prediction results, thereby producing the final prediction results. The proposed method was demonstrated to improve prediction accuracy by 6.4% compared to a conventional neural network.

Using the improved Johnson et al.’s style migration network as a starting point, this paper proposes a new loss function based on the position information Gram matrix. The new method adds the chunked Gram matrix with position information, and simultaneously, the structural similarity between the style map and the resultant image is added to the style training. The style position information is given to the resultant image, and finally, the resolution of the resultant image is improved with the SRGAN. The new model can effectively migrate the texture structure as well as the color space of the style image, while the data of the content image are kept intact. The simulation results reveal that the image processing results of the new model improve those of the classical Johnson et al.’s method, Google Brain team method, and CCPL method, and the SSIM values of the resulting map and style image are all greater than 0.3. As a comparison, the SSIM values of Johnson et al., Google Brain team, and CCPL are 0.14, 0.11, and 0.12, respectively, which is an obvious improvement. Moreover, with deeper training, the new method can improve the similarity of certain resulting images and style images to more than 0.37256. In addition, training other arbitrary content images on the basis of the trained model can quickly yield satisfactory results.

Recently, brain-computer interfaces (BCIs) and brain-machine interfaces have garnered the attention of researchers. Based on connections with external devices, external computers and machines can be controlled by brain signals measured via near-infrared spectroscopy (NIRS) or electroencephalograph devices. Herein, we propose a novel bagging algorithm that generates interpolation data around misclassified data using a possibilistic function, to be applied to BCIs. In contrast to AdaBoost, which is a conventional ensemble learning method that increases the weight of misclassified data to incorporate them with high probability to the next datasets, we generate interpolation data using a membership function centered on misclassified data and incorporate them into the next datasets simultaneously. The interpolated data are known as virtual data herein. By adding the virtual data to the training data, the volume of the training data becomes sufficient for adjusting the discriminate boundary more accurately. Because the membership function is defined as a possibility distribution, this method is named the bagging algorithm based on the possibility distribution. Herein, we formulate a bagging-type ensemble learning based on the possibility distribution and discuss the usefulness of the proposed method for solving simple calculations using NIRS data.

Maintaining high-quality road markings is essential for both safety and traffic flow. However, there has been limited research on automating the process of evaluating the quality of these markings and identifying degraded ones that need to be fixed. Our paper introduces a new approach that uses uncertainty aware (UA) regression to evaluate the quality of road surface markings. The approach is based on deep learning models and a unique training method called “progressive pretraining (PPT).” We used a dataset of RGB images which we converted to binary masks. These masks were then augmented and used to train convolutional neural networks models with a PPT strategy. The results showed that both the hybrid and UA models managed to outperform the baseline model in some metrics such as mean average error which was at 24.38% and accuracy with 81.27%. Additionally, each model showed unique strengths across various performance metrics, highlighting the efficacy of integrating uncertainty and progressive learning in quality assessment tasks. This study presents a solid proof of concept for the application of UA methods in quality evaluation tasks in general, and surface marking quality evaluation in particular.

The pre-chamber pressure is an important control parameter that affects the coke dry quenching process. It often fluctuates violently and is detrimental for the safe operation of the coke dry quenching process. This study proposes an intelligent control method for the pre-chamber pressure based on working condition identification for the coke dry quenching process to realize stable control of the pre-chamber pressure. First, by describing the coke dry quenching process and analyzing the factors affecting the pre-chamber pressure, an intelligent control strategy was developed. Then, the K-means clustering algorithm was used to identify the working conditions of pre-chamber, and the working conditions were divided into two categories: stable and fluctuating. For stable conditions, a fuzzy proportional-integral-derivative controller was designed to improve the pressure control accuracy. For fluctuating conditions, an expert controller was designed to rapidly adjust the pressure. Finally, experiments based on actual data were performed and the results showed that the proposed method can effectively improve the control accuracy of pressure under different conditions. This satisfies the requirements for a continuous coke dry quenching process.

Most of the images encountered in daily life are color images, yet grayscale remains prevalent in many fields due to its reduced data size and simplified operations. When reducing the dimensions of a three-channel color image to a single-channel grayscale, a portion of the original color information is inevitably lost. To yield high-quality grayscale images, this study introduces a color-to-gray conversion method based on chroma difference. This method defines a novel color distance metric for color-to-grayscale conversion, incorporating pixel chromaticity differences alongside brightness variations. The discrepancy in gray pixel values in the output grayscale image effectively mirrors the overall differences among input color image pixels. Optimization is conducted using the conjugate gradient method, ensuring appropriate reflection of luminance information from the input image and chromaticity data from the original color image within the grayscale rendering. Experimental validation confirms the efficacy of this approach. However, as the method accounts for all pixel pairs, it occasionally considers unnecessary pairs, leading to potential distortion in color differences between pixels and consequent inadequacies in chromaticity variation. To address this issue, a weighting factor is introduced, prioritizing color combinations with similar color differences. Experimental findings demonstrate that grayscale images produced using the proposed method outperform those generated by alternative approaches in preserving color differentials and retaining detailed features of the original image. The resulting output exhibits clear and natural outlines, as supported by both subjective and objective assessments.

Pre-trained convolutional neural network (CNN) structures are considered as one of the emerging education management tools that can help improve the quality of education by allowing decision makers to manipulate important indicators. These indicators, which are categorized as student and institution specific factors, may influence student progress, retention or dropout rates. In this paper, we develop a deep learning model of predicting students’ satisfactions and their expected outcomes and associated early failures. The model can also predict dropout rates and identify the main baseline risk factors that influence such rates. The academic data of 12,000 students enrolled from 2018 in the Arab Open University student information system are used as CNNs training dataset to ensure that all institution levels are represented. Then, the trained network provides a probabilistic model that indicates, for each student, the probability of dropout. Based on the prediction model, the study presents an early warning system framework to generate alerts and recommendations to allow early and effective institutional intervention. Experiments are achieved by using the proposed dataset and the performance of our approach is considerably better compared to the competitive models in terms of training/validation accuracy and mean square errors.

In this study, from the perspective of English speech feature parameters, two feature parameters, the mel-frequency cepstral coefficient (MFCC) and filter bank (Fbank), were selected to identify English speech. The algorithms used for recognition employed the classical back-propagation neural network (BPNN), recurrent neural network (RNN), and long short-term memory (LSTM) that were obtained by improving RNN. The three recognition algorithms were compared in the experiments, and the effects of the two feature parameters on the performance of the recognition algorithms were also compared. The LSTM model had the best identification performance among the three neural networks under different experimental environments; the neural network model using the MFCC feature parameter outperformed the neural network using the Fbank feature parameter; the LSTM model had the highest correct rate and the highest speed, while the RNN model ranked second, and the BPNN model ranked worst. The results confirm that the application of the LSTM model in combination with MFCC feature parameter extraction to English speech recognition can achieve higher speech recognition accuracy compared to other neural networks.

Population aging and housing price can affect household consumption, and population aging can indirectly affect household consumption through housing price. This study uses panel data from 31 provinces and cities in China between 2001 and 2018 to construct a panel vector autoregression model for analyzing the impact of population aging and housing price on household consumption level. We determine that population aging and housing price positively impact household consumption level, and population aging also has a positive effect on housing price. The mediating effect model using time series data from 2001 to 2018, with housing price as the mediating variable reveals that population aging directly increases household consumption level, and indirectly increases it through housing price. Finally, this study proposes suggestions such as optimizing the population structure, improving the housing system, and advancing the development of the older adult industry.

The direct optimization of ship hull designs using deep learning algorithms is increasingly expected, as it proposes optimization directions for designers almost instantaneously, without relying on complex, time-consuming, and expensive hydrodynamic simulations. In this study, we proposed a GAN-based 3D ship hull design optimization method. We eliminated the dependence on hydrodynamic simulations by training a separate model to predict ship performance indicators. Instead of a standard discriminator, we applied a relativistic average discriminator to obtain better feedback regarding the anomalous designs. We add two new loss functions for the generator: one restricts design variability, and the other sets improvement targets using feedback from the performance estimation model. In addition, we propose a new training strategy to improve learning effectiveness and avoid instability during training. The experimental results show that our model can optimize the form factor by 5.251% while limiting the deterioration of other indicators and the variability of the ship hull design.

This study uses detailed population statistics and analyzes labor participation rates in China from the perspectives of education and retirement. It presents different hypothetical scenarios and predicts future labor participation rates using queue factors. The results indicate that under the baseline scenario, the overall labor participation rate (51.43%) is projected to significantly decrease by 2060 compared to 2020 (73.76%). The lock-in effect of education leads to a declining participation rate for the 15–24 age group, which persists until approximately the age of 50. Generally, women have higher labor participation rates than men prior to retirement. In the education-centered hypothetical scenario, the quantity impact of educational expansion is evident. Although the relative impact of additional education diminishes toward the end of working life (60–74) compared to the entire working life (15–74). The improvement in the labor market due to educational reform is sustainable across all scenarios. In the retirement-centered hypothetical scenario, reducing retirement rates across age groups increases labor force participation, but this improvement mainly focuses on those under the age of 70 and is not sustained. Thus, delaying retirement policies is only effective in the short term.

This paper uses relevant mathematical statistical models to verify the interactive effect between population mobility and the high-quality development of the manufacturing industry based on the panel data of 43 counties in the Fujian province from 2007 to 2021. The results show that (1) the inflow areas of population flow in the Fujian province are mainly concentrated in the southeast region; (2) there is mutual promotion between population mobility and the high-quality development of the manufacturing industry in the Fujian province, and the population mobility in adjacent areas has a significant promotion effect on the high-quality development of the local manufacturing industry; and (3) heterogeneity results show that after 2011, the interaction effect between population mobility and the high-quality development of the manufacturing industry has declined, and population inflow has a stronger role in promoting the high-quality development of the manufacturing industry than population outflow. The conclusions of this study are of great significance to the Fujian province for implementing a reasonable population policy and promoting the coordinated development of manufacturing enterprises.

This study utilizes a flexible estimation and simulation platform based on the GE-PPML model to analyze the trade and welfare implications of the Regional Comprehensive Economic Partnership (RCEP) for member countries, with a particular focus on China. Our findings suggest that RCEP has the potential to yield significant export gains for member economies. Furthermore, by considering heterogeneity in terms of sectors and product usage, our results indicate that RCEP holds significant potential for deepening value chain trade among member countries, particularly in the manufacturing industry. This study underscores the importance and uniqueness of examining the actual impact of RCEP on member countries’ trade and welfare. By providing clearer insights into these impacts, we aim to contribute to a better understanding of the potential benefits and challenges associated with RCEP implementation.

In this study, the infiltration model was established to study the early warning of pulmonary tuberculosis data in Xiamen public hospitals. Based on the gender characteristics of residents in Xiamen, a percolation model was established to analyze the transmission rates of diseases under different contact types. In addition, the calculation method of the percolation threshold is discussed, and the model is verified by a simulation experiment. The results show that the model can predict the spread of epidemic situations well. The early warning value and relevant preventive measures were obtained by simulating the spread of tuberculosis under different exposure numbers. Bond percolation analysis was used to predict the proportion of the eventually infected population, this threshold of percolation was the basic regeneration number of tuberculosis, and the tuberculosis infection situation was effectively predicted.