An individual’s self-image is a multi-dimensional and multi-structural concept.Its internal dimensions include ability,knowledge,values,personality,and temperament,and its external dimensions are physical appearance,...An individual’s self-image is a multi-dimensional and multi-structural concept.Its internal dimensions include ability,knowledge,values,personality,and temperament,and its external dimensions are physical appearance,behavior,and clothing.A good image will have a positive impact both in life and at work.We will choose appropriate clothing and makeup to modify the external image and cultivate positive qualities such as correct values and an optimistic attitude towards life to enhance internal dimensions.Among them,“personality”and“ability”mostly belong to the research category of mental health education,and“values”fit in the research field of ideological and political education.Ideological and political education and mental health education are both important components of higher education,which show similarities between them.Ideological and political education and mental health education can complement each other in many ways to enhance students’self-image.展开更多
Recent developments in Computer Vision have presented novel opportunities to tackle complex healthcare issues,particularly in the field of lung disease diagnosis.One promising avenue involves the use of chest X-Rays,w...Recent developments in Computer Vision have presented novel opportunities to tackle complex healthcare issues,particularly in the field of lung disease diagnosis.One promising avenue involves the use of chest X-Rays,which are commonly utilized in radiology.To fully exploit their potential,researchers have suggested utilizing deep learning methods to construct computer-aided diagnostic systems.However,constructing and compressing these systems presents a significant challenge,as it relies heavily on the expertise of data scientists.To tackle this issue,we propose an automated approach that utilizes an evolutionary algorithm(EA)to optimize the design and compression of a convolutional neural network(CNN)for X-Ray image classification.Our approach accurately classifies radiography images and detects potential chest abnormalities and infections,including COVID-19.Furthermore,our approach incorporates transfer learning,where a pre-trainedCNNmodel on a vast dataset of chest X-Ray images is fine-tuned for the specific task of detecting COVID-19.This method can help reduce the amount of labeled data required for the task and enhance the overall performance of the model.We have validated our method via a series of experiments against state-of-the-art architectures.展开更多
Accurate identification of rice diseases is crucial for controlling diseases and improving rice yield.To improve the classification accuracy of rice diseases,this paper proposed a classification and identification met...Accurate identification of rice diseases is crucial for controlling diseases and improving rice yield.To improve the classification accuracy of rice diseases,this paper proposed a classification and identification method based on an improved ShuffleNet V2(GE-ShuffleNet)model.Firstly,the Ghost module is used to replace the 1×1 convolution in the two basic unit modules of ShuffleNet V2,and the unimportant 1×1 convolution is deleted from the two basic unit modules of ShuffleNet V2.The Hardswish activation function is applied to replace the ReLU activation function to improve the identification accuracy of the model.Secondly,an effective channel attention(ECA)module is added to the network to avoid dimension reduction,and the correlation between channels is effectively extracted through 1D convolution.Besides,L2 regularization is introduced to fine-tune the training parameters during training to prevent overfitting.Finally,the considerable experimental and numerical results proved the advantages of our proposed model in terms of model size,floating-point operation per second(FLOPs),and parameters(Params).Especially in the case of smaller model size(5.879 M),the identification accuracy of GE-ShuffleNet(96.6%)is higher than that of ShuffleNet V2(94.4%),MobileNet V2(93.7%),AlexNet(79.1%),Swim Transformer(88.1%),EfficientNet V2(89.7%),VGG16(81.9%),GhostNet(89.3%)and ResNet50(92.5%).展开更多
With the application of virtual reality technology to realize interactive display of virtual museum as content of study,we analyze the problems in the current virtual museum system.Taking Daqing Museum for example,we ...With the application of virtual reality technology to realize interactive display of virtual museum as content of study,we analyze the problems in the current virtual museum system.Taking Daqing Museum for example,we develop a 2.5D(between 2D and 3D) architectural modeling technology,and combine it with virtual reality technology,to create the virtual museum simulation platform.By establishing the virtual simulation platform of Daqing Museum,we verify the feasibility of using 2.5D architectural modeling technology to build the virtual museum system,create a virtual simulation platform with practical value,and show the bright future of virtual museum based on 2.5D virtual reality technology.展开更多
Electronic commerce is quickly increasing in several countries, most notably in developing countries. A new electronic-commerce segment known as social commerce has evolved due to the popularity of social media. Consu...Electronic commerce is quickly increasing in several countries, most notably in developing countries. A new electronic-commerce segment known as social commerce has evolved due to the popularity of social media. Consumer trust is important to social commerce success and impacts purchase choices. In modern times, majority of businesses have changed how from the traditional businesses and migrated to social commerce. Electronic commerce was the first of its sort, followed by social commerce, which conducted business via social networking platforms. Identifying the factors that influence social commerce use enables businesses to enhance those features and boost revenue. Thus, the purpose of this study was to examine how increased technology usage influences the social commerce activities of Ghanaian businessmen and women. A review of the literature resulted in the development of a conceptual model. Six hundred and twenty-five responses from Ghanaian enterprises and women who use e-commerce platforms were used to assess the conceptual approach. Partial Least Square Structural Equation Modeling (PLS SEM) was used to validate the model. The reliability and validity of the measuring apparatus were determined using measurement model analysis. To examine the model’s fit and assumptions, we used structural model analysis. Five hypotheses were supported by the structural model data. Effort Expectancy, Perceived Ease of Use, Performance Expectancy, Perceived Utility, and Trust were shown to be the most influential criteria affecting behavioral intention to use s-commerce in Ghana. The findings of this research have major significance for academics and practitioners of social trade.展开更多
Circadian rhythm is an endogenous rhythmic behavior of organisms which can be entrained by daily light–dark cycles.The timing of human sleep-cycle is regulated by endogenous circadian rhythm and homeostatic processes...Circadian rhythm is an endogenous rhythmic behavior of organisms which can be entrained by daily light–dark cycles.The timing of human sleep-cycle is regulated by endogenous circadian rhythm and homeostatic processes. Light exposure affects both sleep timing and circadian rhythm. Now humans can extend lighting time by turning on artificial lights and wake up time is usually triggered by alarm clocks to meet social schedules. This modern lifestyle is believed to be related with a temporal mismatch between sleep and circadian rhythmicity(social jet-lag) and insufficient sleep, which lead to ill mental and physical health outcomes. At present, the impacts of self-selection of light exposure and social constrains on sleep timing is far from clear. According to preferred sleep-wake schedule, there are three different chronotypes. In this paper, we apply a mathematical model to get a quantitative comparison of sleep timing of people with different chronotypes with the effects of modern light consumption and social constrains. The results show that the prolonged day light and evening light exposure both delay preferred sleep timing with the sleep duration almost unchanged. People of evening-type or with longer intrinsic periods are most expected to be vulnerable to evening light. Increasing light exposure can offset the effect of evening light to some extent, but it is most difficult for evening-type people. Social constrains cause the largest social jet-lag in people of evening-type, which increases with evening light intensity or intrinsic periods. Morning-type people's sleep symptoms worsens, while that of evening-type people improves with age. This study provides a theoretical reference for preventing and treating sleep disorder and social jet-lag for individuals with different chronotypes.展开更多
Albacore tuna(Thunnus alalunga)is one of the target species of tuna longline fishing,and waters near the Cook Islands are a vital albacore tuna fishing ground.Marine environmental data are usually presented with diffe...Albacore tuna(Thunnus alalunga)is one of the target species of tuna longline fishing,and waters near the Cook Islands are a vital albacore tuna fishing ground.Marine environmental data are usually presented with different spatial resolutions,which leads to different results in tuna fishery prediction.Study on the impact of different spatial resolutions on the prediction accuracy of albacore tuna fishery to select the best spatial resolution can contribute to better management of albacore tuna resources.The nominal catch per unit effort(CPUE)of albacore tuna is calculated according to vessel monitor system(VMS)data collected from Chinese distantwater fishery enterprises from January 1,2017 to May 31,2021.A total of 26 spatiotemporal and environmental factors,including temperature,salinity,dissolved oxygen of 0–300 m water layer,chlorophyll-a concentration in the sea surface,sea surface height,month,longitude,and latitude,were selected as variables.The temporal resolution of the variables was daily and the spatial resolutions were set to be 0.5°×0.5°,1°×1°,2°×2°,and 5°×5°.The relationship between the nominal CPUE and each individual factor was analyzed to remove the factors irrelavant to the nominal CPUE,together with a multicollinearity diagnosis on the factors to remove factors highly related to the other factors within the four spatial resolutions.The relationship models between CPUE and spatiotemporal and environmental factors by four spatial resolutions were established based on the long short-term memory(LSTM)neural network model.The mean absolute error(MAE)and root mean square error(RMSE)were used to analyze the fitness and accuracy of the models,and to determine the effects of different spatial resolutions on the prediction accuracy of the albacore tuna fishing ground.The results show the resolution of 1°×1°can lead to the best prediction accuracy,with the MAE and RMSE being 0.0268 and 0.0452 respectively,followed by 0.5°×0.5°,2°×2°and 5°×5°with declining prediction accuracy.The results suggested that 1)albacore tuna fishing ground can be predicted by LSTM;2)the VMS records the data in detail and can be used scientifically to calculate the CPUE;3)correlation analysis,and multicollinearity diagnosis are necessary to improve the prediction accuracy of the model;4)the spatial resolution should be 1°×1°in the forecast of albacore tuna fishing ground in waters near the Cook Islands.展开更多
In the emerging field of image segmentation,Fully Convolutional Networks(FCNs)have recently become prominent.However,their effectiveness is intimately linked with the correct selection and fine-tuning of hyperparamete...In the emerging field of image segmentation,Fully Convolutional Networks(FCNs)have recently become prominent.However,their effectiveness is intimately linked with the correct selection and fine-tuning of hyperparameters,which can often be a cumbersome manual task.The main aim of this study is to propose a more efficient,less labour-intensive approach to hyperparameter optimization in FCNs for segmenting fundus images.To this end,our research introduces a hyperparameter-optimized Fully Convolutional Encoder-Decoder Network(FCEDN).The optimization is handled by a novel Genetic Grey Wolf Optimization(G-GWO)algorithm.This algorithm employs the Genetic Algorithm(GA)to generate a diverse set of initial positions.It leverages Grey Wolf Optimization(GWO)to fine-tune these positions within the discrete search space.Testing on the Indian Diabetic Retinopathy Image Dataset(IDRiD),Diabetic Retinopathy,Hypertension,Age-related macular degeneration and Glacuoma ImageS(DR-HAGIS),and Ocular Disease Intelligent Recognition(ODIR)datasets showed that the G-GWO method outperformed four other variants of GWO,GA,and PSO-based hyperparameter optimization techniques.The proposed model achieved impressive segmentation results,with accuracy rates of 98.5%for IDRiD,98.7%for DR-HAGIS,and 98.4%,98.8%,and 98.5%for different sub-datasets within ODIR.These results suggest that the proposed hyperparameter-optimized FCEDN model,driven by the G-GWO algorithm,is more efficient than recent deep-learning models for image segmentation tasks.It thereby presents the potential for increased automation and accuracy in the segmentation of fundus images,mitigating the need for extensive manual hyperparameter adjustments.展开更多
The singular convergence of a chemotaxis-fluid system modeling coral fertilization is justified in spatial dimension three.More precisely,it is shown that a solution of parabolic-parabolic type chemotaxis-fluid system...The singular convergence of a chemotaxis-fluid system modeling coral fertilization is justified in spatial dimension three.More precisely,it is shown that a solution of parabolic-parabolic type chemotaxis-fluid system modeling coral fertilization■converges to that of the parabolic-elliptic type chemotaxis-fluid system modeling coral fertiliz ation■in a certain Fourier-Herz space asε^(-1)→0.展开更多
Federated Learning(FL)enables collaborative and privacy-preserving training of machine learning models within the Internet of Vehicles(IoV)realm.While FL effectively tackles privacy concerns,it also imposes significan...Federated Learning(FL)enables collaborative and privacy-preserving training of machine learning models within the Internet of Vehicles(IoV)realm.While FL effectively tackles privacy concerns,it also imposes significant resource requirements.In traditional FL,trained models are transmitted to a central server for global aggregation,typically in the cloud.This approach often leads to network congestion and bandwidth limitations when numerous devices communicate with the same server.The need for Flexible Global Aggregation and Dynamic Client Selection in FL for the IoV arises from the inherent characteristics of IoV environments.These include diverse and distributed data sources,varying data quality,and limited communication resources.By employing dynamic client selection,we can prioritize relevant and high-quality data sources,enhancing model accuracy.To address this issue,we propose an FL framework that selects global aggregation nodes dynamically rather than a single fixed aggregator.Flexible global aggregation ensures efficient utilization of limited network resources while accommodating the dynamic nature of IoV data sources.This approach optimizes both model performance and resource allocation,making FL in IoV more effective and adaptable.The selection of the global aggregation node is based on workload and communication speed considerations.Additionally,our framework overcomes the constraints associated with network,computational,and energy resources in the IoV environment by implementing a client selection algorithm that dynamically adjusts participants according to predefined parameters.Our approach surpasses Federated Averaging(FedAvg)and Hierarchical FL(HFL)regarding energy consumption,delay,and accuracy,yielding superior results.展开更多
The performance of Wireless Sensor Networks(WSNs)is an important fragment of the Internet of Things(IoT),where the current WSNbuilt IoT network’s sensor hubs are enticing due to their critical resources.By grouping h...The performance of Wireless Sensor Networks(WSNs)is an important fragment of the Internet of Things(IoT),where the current WSNbuilt IoT network’s sensor hubs are enticing due to their critical resources.By grouping hubs,a clustering convention offers a useful solution for ensuring energy-saving of hubs andHybridMedia Access Control(HMAC)during the course of the organization.Nevertheless,current grouping standards suffer from issues with the grouping structure that impacts the exhibition of these conventions negatively.In this investigation,we recommend an Improved Energy-Proficient Algorithm(IEPA)for HMAC throughout the lifetime of the WSN-based IoT.Three consecutive segments are suggested.For the covering of adjusted clusters,an ideal number of clusters is determined first.Then,fair static clusters are shaped,based on an updated calculation for fluffy cluster heads,to reduce and adapt the energy use of the sensor hubs.Cluster heads(CHs)are,ultimately,selected in optimal locations,with the pivot of the cluster heads working among cluster members.Specifically,the proposed convention diminishes and balances the energy utilization of hubs by improving the grouping structure,where the IEPAis reasonable for systems that need a long time.The assessment results demonstrate that the IEPA performs better than existing conventions.展开更多
As the amount of medical images transmitted over networks and kept on online servers continues to rise,the need to protect those images digitally is becoming increasingly important.However,due to the massive amounts o...As the amount of medical images transmitted over networks and kept on online servers continues to rise,the need to protect those images digitally is becoming increasingly important.However,due to the massive amounts of multimedia and medical pictures being exchanged,low computational complexity techniques have been developed.Most commonly used algorithms offer very little security and require a great deal of communication,all of which add to the high processing costs associated with using them.First,a deep learning classifier is used to classify records according to the degree of concealment they require.Medical images that aren’t needed can be saved by using this method,which cuts down on security costs.Encryption is one of the most effective methods for protecting medical images after this step.Confusion and dispersion are two fundamental encryption processes.A new encryption algorithm for very sensitive data is developed in this study.Picture splitting with image blocks is nowdeveloped by using Zigzag patterns,rotation of the image blocks,and random permutation for scrambling the blocks.After that,this research suggests a Region of Interest(ROI)technique based on selective picture encryption.For the first step,we use an active contour picture segmentation to separate the ROI from the Region of Background(ROB).Permutation and diffusion are then carried out using a Hilbert curve and a Skew Tent map.Once all of the blocks have been encrypted,they are combined to create encrypted images.The investigational analysis is carried out to test the competence of the projected ideal with existing techniques.展开更多
Developments in biomedical science, signal processing technologies have led Electroencephalography (EEG) signals to be widely used in the diagnosis of brain disease and in the field of Brain-Computer Interface (BCI). ...Developments in biomedical science, signal processing technologies have led Electroencephalography (EEG) signals to be widely used in the diagnosis of brain disease and in the field of Brain-Computer Interface (BCI). The collected EEG signals are processed using Machine Learning-Random Forest and Naive Bayes- and Deep Learning-Recurrent Neural Network (RNN), Neural Network (NN) and Long Short Term Memory (LSTM)-Algorithms to obtain the recent mood of a person. The Algorithms mentioned above have been imposed on the data set in order to find out what the person is feeling at a particular moment. The following thesis is conducted to find out one of the following moods (happy, surprised, disgust, fear, anger and sadness) of a person at an instant, with an aim to obtain the result with least amount of time delay as the mood differs. It is pretty obvious that the accuracy of the output varies depending upon the algorithm used, time taken to process the data, so that it is easy for us to compare the reliability and dependency of a particular algorithm to another, prior to its practical implementation. The imbalance data sets that were used had an imbalanced class and thus, over fitting occurred. This problem was handled by generating Artificial Data sets with the use of SMOTE Oversampling Technique.展开更多
The Chinese paper-cut art,first recorded in the Wei,Jin,and Southern and Northern Dynasties(220 AD-589 AD),has witnessed the changes of times,yet it still retains its artistic vitality.Chinese papercuts can be divided...The Chinese paper-cut art,first recorded in the Wei,Jin,and Southern and Northern Dynasties(220 AD-589 AD),has witnessed the changes of times,yet it still retains its artistic vitality.Chinese papercuts can be divided into two schools:the northern and the southern.Jiangsu,located in the region of the Yellow River and Huai River,is the geographical dividing line between those two schools.Therefore,in Jiangsu Province,not only the rough northern art form(such as in Xuzhou papercut)but also the graceful southern art form(such as in Jintan papercut)is evident.In addition,the unique combined paper-cut style(such as in Yangzhou and Nanjing papercuts)can be appreciated here.Although several scholars have analyzed the artistic characteristics of Jiangsu papercut based on cultural background,very few have discussed the differences between the northern and the southern in terms of content,connotation,and style.Through literature review and collected works made by local craftsmen and inheritors of this tradition,this article aims to show readers the contrast and integration of papercuts in these four places under the influence of different cultural and economic backgrounds in order to better understand the role of regional factors in shaping the art form of papercuts in Jiangsu Province.Nowadays,with the change in people’s lifestyles,the living space of traditional papercuts has shrunk drastically,and its practicability in the past has faded.Instead,people are searching for and creating cultural and artistic value in museums,tourist attractions,and commodity transactions.Among them,some works have deviated from the cultural background of traditional paper-cut art.Therefore,this paper provides a basis for the current development of this art form in Jiangsu.展开更多
To improve the processing efficiency and the quality of orbital milling hole of aerospace Al-alloy, the big-pitch influence on cutting force and hole quality was studied experimentally. First, a program based on horiz...To improve the processing efficiency and the quality of orbital milling hole of aerospace Al-alloy, the big-pitch influence on cutting force and hole quality was studied experimentally. First, a program based on horizontal lathe was proposed based on kinematics analysis of orbital milling. Then, the cutting force at different stages and the hole quality with different pitches were measured. Results show that the axial force and radial force increase with the pitch amplification during orbital milling. However, the axial force in the orbital milling hole is about 8—10 times smaller than that in the conventional drilling. The diameter error of milling hole is 48—93 μm, and the surface roughness of milling hole is 1.2—1.7 μm. Finally, an orbital milling device with big pitch was designed.展开更多
Heterogeneous wireless access technologies will coexist in next generation wireless networks.These technologies form integrated networks,and these networks support multiple services with high quality level.Various acc...Heterogeneous wireless access technologies will coexist in next generation wireless networks.These technologies form integrated networks,and these networks support multiple services with high quality level.Various access technologies allow users to select the best available access network to meet the requirements of each type of communication service.Being always best connected anytime and anywhere is a major concern in a heterogeneous wireless networks environment.Always best connected enables network selection mechanisms to keep mobile users always connected to the best network.We present an overview of the network selection and prediction problems and challenges.In addition,we discuss a comprehensive classification of related theoretic approaches,and also study the integration between these methods,finding the best solution of network selection and prediction problems.The optimal solution can fulfill the requirements of the next generation wireless networks.展开更多
To avoid the machine problems of excessive axial force, complex process flow and frequent tool changing during robotic drilling holes, a new hole-making technology (i.e., helical milling hole) was introduced for desig...To avoid the machine problems of excessive axial force, complex process flow and frequent tool changing during robotic drilling holes, a new hole-making technology (i.e., helical milling hole) was introduced for designing a new robotic helical milling hole system, which could further improve robotic hole-making ability in airplane digital assembly. After analysis on the characteristics of helical milling hole, advantages and limitations of two typical robotic helical milling hole systems were summarized. Then, vector model of helical milling hole movement was built on vector analysis method. Finally, surface roughness calculation formula was deduced according to the movement principle of helical milling hole, then the influence of main technological parameters on surface roughness was analyzed. Analysis shows that theoretical surface roughness of hole becomes poor with the increase of tool speed ratio and revolution radius. Meanwhile, the roughness decreases according to the increase of tool teeth number. The research contributes greatly to the construction of roughness prediction model in helical milling hole.展开更多
To solve the problem of difficulty in identifying apple diseases in the natural environment and the low application rate of deep learning recognition networks,a lightweight ResNet(LW-ResNet)model for apple disease rec...To solve the problem of difficulty in identifying apple diseases in the natural environment and the low application rate of deep learning recognition networks,a lightweight ResNet(LW-ResNet)model for apple disease recognition is proposed.Based on the deep residual network(ResNet18),the multi-scale feature extraction layer is constructed by group convolution to realize the compression model and improve the extraction ability of different sizes of lesion features.By improving the identity mapping structure to reduce information loss.By introducing the efficient channel attention module(ECANet)to suppress noise from a complex background.The experimental results show that the average precision,recall and F1-score of the LW-ResNet on the test set are 97.80%,97.92%and 97.85%,respectively.The parameter memory is 2.32 MB,which is 94%less than that of ResNet18.Compared with the classic lightweight networks SqueezeNet and MobileNetV2,LW-ResNet has obvious advantages in recognition performance,speed,parameter memory requirement and time complexity.The proposed model has the advantages of low computational cost,low storage cost,strong real-time performance,high identification accuracy,and strong practicability,which can meet the needs of real-time identification task of apple leaf disease on resource-constrained devices.展开更多
To implement the computation of AHP (Analytic Hierarchy Process) automatically, make the implementation inde-pendence on the platform, and to convenient the service composi-tion and service computation, this study imp...To implement the computation of AHP (Analytic Hierarchy Process) automatically, make the implementation inde-pendence on the platform, and to convenient the service composi-tion and service computation, this study implemented AHP model as a Web service and stored the evaluation information as XML form. This paper introduces the overall demands of AHP evalua-tion system firstly, and then presents some key technologies to implement AHP evaluation system, including mainly AHP evalua-tion index architecture based on XML, the class of AHP model, the Web service encapsulation of AHP, the publication and invoca-tion of AHP and the client application. At last, the system is used to analyze the impact factors of online consumption behavior.展开更多
基金Sichuan Provincial Key Research Base for Philosophy and Social Sciences,Mental Health Education Research Project“Research on Self-Image Cognition Improvement Strategies of Higher Vocational Students Under the Background of Ideological and Political Education in Colleges and Universities”(XLJKJY202114C)。
文摘An individual’s self-image is a multi-dimensional and multi-structural concept.Its internal dimensions include ability,knowledge,values,personality,and temperament,and its external dimensions are physical appearance,behavior,and clothing.A good image will have a positive impact both in life and at work.We will choose appropriate clothing and makeup to modify the external image and cultivate positive qualities such as correct values and an optimistic attitude towards life to enhance internal dimensions.Among them,“personality”and“ability”mostly belong to the research category of mental health education,and“values”fit in the research field of ideological and political education.Ideological and political education and mental health education are both important components of higher education,which show similarities between them.Ideological and political education and mental health education can complement each other in many ways to enhance students’self-image.
基金via funding from Prince Sattam bin Abdulaziz University Project Number(PSAU/2023/R/1444).
文摘Recent developments in Computer Vision have presented novel opportunities to tackle complex healthcare issues,particularly in the field of lung disease diagnosis.One promising avenue involves the use of chest X-Rays,which are commonly utilized in radiology.To fully exploit their potential,researchers have suggested utilizing deep learning methods to construct computer-aided diagnostic systems.However,constructing and compressing these systems presents a significant challenge,as it relies heavily on the expertise of data scientists.To tackle this issue,we propose an automated approach that utilizes an evolutionary algorithm(EA)to optimize the design and compression of a convolutional neural network(CNN)for X-Ray image classification.Our approach accurately classifies radiography images and detects potential chest abnormalities and infections,including COVID-19.Furthermore,our approach incorporates transfer learning,where a pre-trainedCNNmodel on a vast dataset of chest X-Ray images is fine-tuned for the specific task of detecting COVID-19.This method can help reduce the amount of labeled data required for the task and enhance the overall performance of the model.We have validated our method via a series of experiments against state-of-the-art architectures.
基金This work is supported in part by the Ji Lin provincial science and technology department international science and technology cooperation project under Grant 20200801014GHthe Changchun City Science and Technology Bureau key science and technology research projects under Grant 21ZGN28.
文摘Accurate identification of rice diseases is crucial for controlling diseases and improving rice yield.To improve the classification accuracy of rice diseases,this paper proposed a classification and identification method based on an improved ShuffleNet V2(GE-ShuffleNet)model.Firstly,the Ghost module is used to replace the 1×1 convolution in the two basic unit modules of ShuffleNet V2,and the unimportant 1×1 convolution is deleted from the two basic unit modules of ShuffleNet V2.The Hardswish activation function is applied to replace the ReLU activation function to improve the identification accuracy of the model.Secondly,an effective channel attention(ECA)module is added to the network to avoid dimension reduction,and the correlation between channels is effectively extracted through 1D convolution.Besides,L2 regularization is introduced to fine-tune the training parameters during training to prevent overfitting.Finally,the considerable experimental and numerical results proved the advantages of our proposed model in terms of model size,floating-point operation per second(FLOPs),and parameters(Params).Especially in the case of smaller model size(5.879 M),the identification accuracy of GE-ShuffleNet(96.6%)is higher than that of ShuffleNet V2(94.4%),MobileNet V2(93.7%),AlexNet(79.1%),Swim Transformer(88.1%),EfficientNet V2(89.7%),VGG16(81.9%),GhostNet(89.3%)and ResNet50(92.5%).
文摘With the application of virtual reality technology to realize interactive display of virtual museum as content of study,we analyze the problems in the current virtual museum system.Taking Daqing Museum for example,we develop a 2.5D(between 2D and 3D) architectural modeling technology,and combine it with virtual reality technology,to create the virtual museum simulation platform.By establishing the virtual simulation platform of Daqing Museum,we verify the feasibility of using 2.5D architectural modeling technology to build the virtual museum system,create a virtual simulation platform with practical value,and show the bright future of virtual museum based on 2.5D virtual reality technology.
文摘Electronic commerce is quickly increasing in several countries, most notably in developing countries. A new electronic-commerce segment known as social commerce has evolved due to the popularity of social media. Consumer trust is important to social commerce success and impacts purchase choices. In modern times, majority of businesses have changed how from the traditional businesses and migrated to social commerce. Electronic commerce was the first of its sort, followed by social commerce, which conducted business via social networking platforms. Identifying the factors that influence social commerce use enables businesses to enhance those features and boost revenue. Thus, the purpose of this study was to examine how increased technology usage influences the social commerce activities of Ghanaian businessmen and women. A review of the literature resulted in the development of a conceptual model. Six hundred and twenty-five responses from Ghanaian enterprises and women who use e-commerce platforms were used to assess the conceptual approach. Partial Least Square Structural Equation Modeling (PLS SEM) was used to validate the model. The reliability and validity of the measuring apparatus were determined using measurement model analysis. To examine the model’s fit and assumptions, we used structural model analysis. Five hypotheses were supported by the structural model data. Effort Expectancy, Perceived Ease of Use, Performance Expectancy, Perceived Utility, and Trust were shown to be the most influential criteria affecting behavioral intention to use s-commerce in Ghana. The findings of this research have major significance for academics and practitioners of social trade.
基金Project supported by the National Natural Science Foundation of China (Grant No. 11672177)。
文摘Circadian rhythm is an endogenous rhythmic behavior of organisms which can be entrained by daily light–dark cycles.The timing of human sleep-cycle is regulated by endogenous circadian rhythm and homeostatic processes. Light exposure affects both sleep timing and circadian rhythm. Now humans can extend lighting time by turning on artificial lights and wake up time is usually triggered by alarm clocks to meet social schedules. This modern lifestyle is believed to be related with a temporal mismatch between sleep and circadian rhythmicity(social jet-lag) and insufficient sleep, which lead to ill mental and physical health outcomes. At present, the impacts of self-selection of light exposure and social constrains on sleep timing is far from clear. According to preferred sleep-wake schedule, there are three different chronotypes. In this paper, we apply a mathematical model to get a quantitative comparison of sleep timing of people with different chronotypes with the effects of modern light consumption and social constrains. The results show that the prolonged day light and evening light exposure both delay preferred sleep timing with the sleep duration almost unchanged. People of evening-type or with longer intrinsic periods are most expected to be vulnerable to evening light. Increasing light exposure can offset the effect of evening light to some extent, but it is most difficult for evening-type people. Social constrains cause the largest social jet-lag in people of evening-type, which increases with evening light intensity or intrinsic periods. Morning-type people's sleep symptoms worsens, while that of evening-type people improves with age. This study provides a theoretical reference for preventing and treating sleep disorder and social jet-lag for individuals with different chronotypes.
基金the National Natural Science Foundation of China(No.32273185)the National Key R&D Program of China(No.2020YFD0901205)the Marine Fishery Resources Investigation and Exploration Program of the Ministry of Agriculture and Rural Affairs of China in 2021(No.D-8006-21-0215)。
文摘Albacore tuna(Thunnus alalunga)is one of the target species of tuna longline fishing,and waters near the Cook Islands are a vital albacore tuna fishing ground.Marine environmental data are usually presented with different spatial resolutions,which leads to different results in tuna fishery prediction.Study on the impact of different spatial resolutions on the prediction accuracy of albacore tuna fishery to select the best spatial resolution can contribute to better management of albacore tuna resources.The nominal catch per unit effort(CPUE)of albacore tuna is calculated according to vessel monitor system(VMS)data collected from Chinese distantwater fishery enterprises from January 1,2017 to May 31,2021.A total of 26 spatiotemporal and environmental factors,including temperature,salinity,dissolved oxygen of 0–300 m water layer,chlorophyll-a concentration in the sea surface,sea surface height,month,longitude,and latitude,were selected as variables.The temporal resolution of the variables was daily and the spatial resolutions were set to be 0.5°×0.5°,1°×1°,2°×2°,and 5°×5°.The relationship between the nominal CPUE and each individual factor was analyzed to remove the factors irrelavant to the nominal CPUE,together with a multicollinearity diagnosis on the factors to remove factors highly related to the other factors within the four spatial resolutions.The relationship models between CPUE and spatiotemporal and environmental factors by four spatial resolutions were established based on the long short-term memory(LSTM)neural network model.The mean absolute error(MAE)and root mean square error(RMSE)were used to analyze the fitness and accuracy of the models,and to determine the effects of different spatial resolutions on the prediction accuracy of the albacore tuna fishing ground.The results show the resolution of 1°×1°can lead to the best prediction accuracy,with the MAE and RMSE being 0.0268 and 0.0452 respectively,followed by 0.5°×0.5°,2°×2°and 5°×5°with declining prediction accuracy.The results suggested that 1)albacore tuna fishing ground can be predicted by LSTM;2)the VMS records the data in detail and can be used scientifically to calculate the CPUE;3)correlation analysis,and multicollinearity diagnosis are necessary to improve the prediction accuracy of the model;4)the spatial resolution should be 1°×1°in the forecast of albacore tuna fishing ground in waters near the Cook Islands.
基金supported in part by the National Natural Science Foundation of China under Grant 11527801 and 41706201.
文摘In the emerging field of image segmentation,Fully Convolutional Networks(FCNs)have recently become prominent.However,their effectiveness is intimately linked with the correct selection and fine-tuning of hyperparameters,which can often be a cumbersome manual task.The main aim of this study is to propose a more efficient,less labour-intensive approach to hyperparameter optimization in FCNs for segmenting fundus images.To this end,our research introduces a hyperparameter-optimized Fully Convolutional Encoder-Decoder Network(FCEDN).The optimization is handled by a novel Genetic Grey Wolf Optimization(G-GWO)algorithm.This algorithm employs the Genetic Algorithm(GA)to generate a diverse set of initial positions.It leverages Grey Wolf Optimization(GWO)to fine-tune these positions within the discrete search space.Testing on the Indian Diabetic Retinopathy Image Dataset(IDRiD),Diabetic Retinopathy,Hypertension,Age-related macular degeneration and Glacuoma ImageS(DR-HAGIS),and Ocular Disease Intelligent Recognition(ODIR)datasets showed that the G-GWO method outperformed four other variants of GWO,GA,and PSO-based hyperparameter optimization techniques.The proposed model achieved impressive segmentation results,with accuracy rates of 98.5%for IDRiD,98.7%for DR-HAGIS,and 98.4%,98.8%,and 98.5%for different sub-datasets within ODIR.These results suggest that the proposed hyperparameter-optimized FCEDN model,driven by the G-GWO algorithm,is more efficient than recent deep-learning models for image segmentation tasks.It thereby presents the potential for increased automation and accuracy in the segmentation of fundus images,mitigating the need for extensive manual hyperparameter adjustments.
基金Supported by the NSFC(12161041,12001435 and12071197)the training program for academic and technical leaders of major disciplines in Jiangxi Province(20204BCJL23057)+2 种基金the Natural Science Foundation of Jiangxi Province(20212BAB201008)the Educational Commission Science Programm of Jiangxi Province(GJJ190272)Natural Science Foundation of Shandong Province(ZR2021MA031)。
文摘The singular convergence of a chemotaxis-fluid system modeling coral fertilization is justified in spatial dimension three.More precisely,it is shown that a solution of parabolic-parabolic type chemotaxis-fluid system modeling coral fertilization■converges to that of the parabolic-elliptic type chemotaxis-fluid system modeling coral fertiliz ation■in a certain Fourier-Herz space asε^(-1)→0.
基金supported by the UAE University UPAR Research Grant Program under Grant 31T122.
文摘Federated Learning(FL)enables collaborative and privacy-preserving training of machine learning models within the Internet of Vehicles(IoV)realm.While FL effectively tackles privacy concerns,it also imposes significant resource requirements.In traditional FL,trained models are transmitted to a central server for global aggregation,typically in the cloud.This approach often leads to network congestion and bandwidth limitations when numerous devices communicate with the same server.The need for Flexible Global Aggregation and Dynamic Client Selection in FL for the IoV arises from the inherent characteristics of IoV environments.These include diverse and distributed data sources,varying data quality,and limited communication resources.By employing dynamic client selection,we can prioritize relevant and high-quality data sources,enhancing model accuracy.To address this issue,we propose an FL framework that selects global aggregation nodes dynamically rather than a single fixed aggregator.Flexible global aggregation ensures efficient utilization of limited network resources while accommodating the dynamic nature of IoV data sources.This approach optimizes both model performance and resource allocation,making FL in IoV more effective and adaptable.The selection of the global aggregation node is based on workload and communication speed considerations.Additionally,our framework overcomes the constraints associated with network,computational,and energy resources in the IoV environment by implementing a client selection algorithm that dynamically adjusts participants according to predefined parameters.Our approach surpasses Federated Averaging(FedAvg)and Hierarchical FL(HFL)regarding energy consumption,delay,and accuracy,yielding superior results.
文摘The performance of Wireless Sensor Networks(WSNs)is an important fragment of the Internet of Things(IoT),where the current WSNbuilt IoT network’s sensor hubs are enticing due to their critical resources.By grouping hubs,a clustering convention offers a useful solution for ensuring energy-saving of hubs andHybridMedia Access Control(HMAC)during the course of the organization.Nevertheless,current grouping standards suffer from issues with the grouping structure that impacts the exhibition of these conventions negatively.In this investigation,we recommend an Improved Energy-Proficient Algorithm(IEPA)for HMAC throughout the lifetime of the WSN-based IoT.Three consecutive segments are suggested.For the covering of adjusted clusters,an ideal number of clusters is determined first.Then,fair static clusters are shaped,based on an updated calculation for fluffy cluster heads,to reduce and adapt the energy use of the sensor hubs.Cluster heads(CHs)are,ultimately,selected in optimal locations,with the pivot of the cluster heads working among cluster members.Specifically,the proposed convention diminishes and balances the energy utilization of hubs by improving the grouping structure,where the IEPAis reasonable for systems that need a long time.The assessment results demonstrate that the IEPA performs better than existing conventions.
文摘As the amount of medical images transmitted over networks and kept on online servers continues to rise,the need to protect those images digitally is becoming increasingly important.However,due to the massive amounts of multimedia and medical pictures being exchanged,low computational complexity techniques have been developed.Most commonly used algorithms offer very little security and require a great deal of communication,all of which add to the high processing costs associated with using them.First,a deep learning classifier is used to classify records according to the degree of concealment they require.Medical images that aren’t needed can be saved by using this method,which cuts down on security costs.Encryption is one of the most effective methods for protecting medical images after this step.Confusion and dispersion are two fundamental encryption processes.A new encryption algorithm for very sensitive data is developed in this study.Picture splitting with image blocks is nowdeveloped by using Zigzag patterns,rotation of the image blocks,and random permutation for scrambling the blocks.After that,this research suggests a Region of Interest(ROI)technique based on selective picture encryption.For the first step,we use an active contour picture segmentation to separate the ROI from the Region of Background(ROB).Permutation and diffusion are then carried out using a Hilbert curve and a Skew Tent map.Once all of the blocks have been encrypted,they are combined to create encrypted images.The investigational analysis is carried out to test the competence of the projected ideal with existing techniques.
文摘Developments in biomedical science, signal processing technologies have led Electroencephalography (EEG) signals to be widely used in the diagnosis of brain disease and in the field of Brain-Computer Interface (BCI). The collected EEG signals are processed using Machine Learning-Random Forest and Naive Bayes- and Deep Learning-Recurrent Neural Network (RNN), Neural Network (NN) and Long Short Term Memory (LSTM)-Algorithms to obtain the recent mood of a person. The Algorithms mentioned above have been imposed on the data set in order to find out what the person is feeling at a particular moment. The following thesis is conducted to find out one of the following moods (happy, surprised, disgust, fear, anger and sadness) of a person at an instant, with an aim to obtain the result with least amount of time delay as the mood differs. It is pretty obvious that the accuracy of the output varies depending upon the algorithm used, time taken to process the data, so that it is easy for us to compare the reliability and dependency of a particular algorithm to another, prior to its practical implementation. The imbalance data sets that were used had an imbalanced class and thus, over fitting occurred. This problem was handled by generating Artificial Data sets with the use of SMOTE Oversampling Technique.
文摘The Chinese paper-cut art,first recorded in the Wei,Jin,and Southern and Northern Dynasties(220 AD-589 AD),has witnessed the changes of times,yet it still retains its artistic vitality.Chinese papercuts can be divided into two schools:the northern and the southern.Jiangsu,located in the region of the Yellow River and Huai River,is the geographical dividing line between those two schools.Therefore,in Jiangsu Province,not only the rough northern art form(such as in Xuzhou papercut)but also the graceful southern art form(such as in Jintan papercut)is evident.In addition,the unique combined paper-cut style(such as in Yangzhou and Nanjing papercuts)can be appreciated here.Although several scholars have analyzed the artistic characteristics of Jiangsu papercut based on cultural background,very few have discussed the differences between the northern and the southern in terms of content,connotation,and style.Through literature review and collected works made by local craftsmen and inheritors of this tradition,this article aims to show readers the contrast and integration of papercuts in these four places under the influence of different cultural and economic backgrounds in order to better understand the role of regional factors in shaping the art form of papercuts in Jiangsu Province.Nowadays,with the change in people’s lifestyles,the living space of traditional papercuts has shrunk drastically,and its practicability in the past has faded.Instead,people are searching for and creating cultural and artistic value in museums,tourist attractions,and commodity transactions.Among them,some works have deviated from the cultural background of traditional paper-cut art.Therefore,this paper provides a basis for the current development of this art form in Jiangsu.
基金Supported by National Natural Science Foundation of China (No.50975141 and No.51005118)Aviation Science Fund (No.20091652018 and No.2010352005)
文摘To improve the processing efficiency and the quality of orbital milling hole of aerospace Al-alloy, the big-pitch influence on cutting force and hole quality was studied experimentally. First, a program based on horizontal lathe was proposed based on kinematics analysis of orbital milling. Then, the cutting force at different stages and the hole quality with different pitches were measured. Results show that the axial force and radial force increase with the pitch amplification during orbital milling. However, the axial force in the orbital milling hole is about 8—10 times smaller than that in the conventional drilling. The diameter error of milling hole is 48—93 μm, and the surface roughness of milling hole is 1.2—1.7 μm. Finally, an orbital milling device with big pitch was designed.
基金funded by the University of Malaya, under Grant No.RG208-11AFR
文摘Heterogeneous wireless access technologies will coexist in next generation wireless networks.These technologies form integrated networks,and these networks support multiple services with high quality level.Various access technologies allow users to select the best available access network to meet the requirements of each type of communication service.Being always best connected anytime and anywhere is a major concern in a heterogeneous wireless networks environment.Always best connected enables network selection mechanisms to keep mobile users always connected to the best network.We present an overview of the network selection and prediction problems and challenges.In addition,we discuss a comprehensive classification of related theoretic approaches,and also study the integration between these methods,finding the best solution of network selection and prediction problems.The optimal solution can fulfill the requirements of the next generation wireless networks.
基金Foundation item: Projects(50975141, 51005118) supported by the National Natural Science Foundation of China Projects(20091652018, 2010352005) supported by Aviation Science Fund of China Project(YKJ11-001) supported by Key Program of Nanjing College of Information Technology Institute, China
文摘To avoid the machine problems of excessive axial force, complex process flow and frequent tool changing during robotic drilling holes, a new hole-making technology (i.e., helical milling hole) was introduced for designing a new robotic helical milling hole system, which could further improve robotic hole-making ability in airplane digital assembly. After analysis on the characteristics of helical milling hole, advantages and limitations of two typical robotic helical milling hole systems were summarized. Then, vector model of helical milling hole movement was built on vector analysis method. Finally, surface roughness calculation formula was deduced according to the movement principle of helical milling hole, then the influence of main technological parameters on surface roughness was analyzed. Analysis shows that theoretical surface roughness of hole becomes poor with the increase of tool speed ratio and revolution radius. Meanwhile, the roughness decreases according to the increase of tool teeth number. The research contributes greatly to the construction of roughness prediction model in helical milling hole.
基金funded by the Science and Technology Development Program of Jilin Province(20190301024NY)the Precision Agriculture and Big Data Engineering Research Center of Jilin Province(2020C005).
文摘To solve the problem of difficulty in identifying apple diseases in the natural environment and the low application rate of deep learning recognition networks,a lightweight ResNet(LW-ResNet)model for apple disease recognition is proposed.Based on the deep residual network(ResNet18),the multi-scale feature extraction layer is constructed by group convolution to realize the compression model and improve the extraction ability of different sizes of lesion features.By improving the identity mapping structure to reduce information loss.By introducing the efficient channel attention module(ECANet)to suppress noise from a complex background.The experimental results show that the average precision,recall and F1-score of the LW-ResNet on the test set are 97.80%,97.92%and 97.85%,respectively.The parameter memory is 2.32 MB,which is 94%less than that of ResNet18.Compared with the classic lightweight networks SqueezeNet and MobileNetV2,LW-ResNet has obvious advantages in recognition performance,speed,parameter memory requirement and time complexity.The proposed model has the advantages of low computational cost,low storage cost,strong real-time performance,high identification accuracy,and strong practicability,which can meet the needs of real-time identification task of apple leaf disease on resource-constrained devices.
基金the National Natural Science Foundation of China(70772073)the Natural Science Foundation of Shanghai (07ZR14003)+1 种基金the Social Science Programming Foundation of Shanghai(2007BZH001)Excellent Young Teachers Follow-Up Scientific Research Foundation of Shanghai Fisheries University(A-0212-07-0132)
文摘To implement the computation of AHP (Analytic Hierarchy Process) automatically, make the implementation inde-pendence on the platform, and to convenient the service composi-tion and service computation, this study implemented AHP model as a Web service and stored the evaluation information as XML form. This paper introduces the overall demands of AHP evalua-tion system firstly, and then presents some key technologies to implement AHP evaluation system, including mainly AHP evalua-tion index architecture based on XML, the class of AHP model, the Web service encapsulation of AHP, the publication and invoca-tion of AHP and the client application. At last, the system is used to analyze the impact factors of online consumption behavior.