OBJECTIVE:To develop an automated system for identifying and classifying constitution types in Traditional Chinese Medicine(TCM)by leveraging multi-model fusion algorithms.METHODS:A condensed version of a physical inf...OBJECTIVE:To develop an automated system for identifying and classifying constitution types in Traditional Chinese Medicine(TCM)by leveraging multi-model fusion algorithms.METHODS:A condensed version of a physical information collection form was designed to facilitate efficient data acquisition.The collected data were analyzed using a multi-model fusion approach,which integrated several machine learning techniques.These included support vector machines,Naive Bayes,decision trees,random forests,logistic regression,multilayer perceptrons,K-nearest neighbors,gradient boosting,adaptive ensemble learning,and recurrent neural networks.A soft voting strategy was used to combine the predictive outputs of each model,enabling the selection of the most effective model combination.RESULTS:The classification models demonstrated consistent and robust performance across most TCM constitution types when enhanced by the multi-model fusion strategy.In particular,high levels of accuracy,precision,recall,and F1-score were achieved for constitution types such as Yang deficiency,Qi deficiency,and Qi stagnation.However,the classification performance for the Yin deficiency constitution was relatively lower,indicating the need for further refinement and optimization in future research.CONCLUSION:This study introduces a novel,automated method for classifying TCM constitution types through the application of multi-model fusion algorithms.The approach simplifies the complex task of constitution identification while offering a practical and theoretical framework for the intelligent diagnosis of TCM body types.The findings have the potential to enhance personalized health management and support clinical decision-making in TCM diagnosis and treatment.展开更多
[Objectives]This study was conducted to achieve rapid and accurate detection of protein content in rice with a particle size of 1.0 mm.[Methods]A multi-model fusion strategy was proposed on the basis of Stacking ensem...[Objectives]This study was conducted to achieve rapid and accurate detection of protein content in rice with a particle size of 1.0 mm.[Methods]A multi-model fusion strategy was proposed on the basis of Stacking ensemble learning.A base learner pool was constructed,containing Partial Least Squares(PLS),Support Vector Machine(SVM),Deep Extreme Learning Machine(DELM),Random Forest(RF),Gradient Boosting Decision Tree(GBDT),and Multilayer Perceptron(MLP).PLS,DELM,and Linear Regression(LR)were used as meta-learner candidates.Employing integer coding technology,systematic dynamic combinations of base learners and meta-learners were generated,resulting in a total of 40 non-repetitive fusion models.The optimal combination was selected through a comprehensive evaluation based on multiple assessment indicators.[Results]The combination"PLS-DELM-MLP-LR"(code 1367)achieved coefficients of determination of 0.9732 and 0.9780 on the validation set and independent test set,respectively,with relative root mean square errors of 2.35%and 2.36%,and residual predictive deviations of 6.1075 and 6.7479,respectively.[Conclusions]The Stacking fusion model significantly enhances the predictive accuracy and robustness of spectral quantitative analysis,providing an efficient and feasible solution for modeling complex agricultural product spectral data.展开更多
This paper addresses the accuracy and timeliness limitations of traditional comprehensive prediction methods by proposing an approach of decision-level fusion of multisource data.A risk prediction indicator system was...This paper addresses the accuracy and timeliness limitations of traditional comprehensive prediction methods by proposing an approach of decision-level fusion of multisource data.A risk prediction indicator system was established for water and mud inrush in tunnels by analyzing advanced prediction data for specifi c tunnel segments.Additionally,the indicator weights were determined using the analytic hierarchy process combined with the Huber weighting method.Subsequently,a multisource data decision-layer fusion algorithm was utilized to generate fused imaging results for tunnel water and mud inrush risk predictions.Meanwhile,risk analysis was performed for different tunnel sections to achieve spatial and temporal complementarity within the indicator system and optimize redundant information.Finally,model feasibility was validated using the CZ Project Sejila Mountain Tunnel segment as a case study,yielding favorable risk prediction results and enabling effi cient information fusion and support for construction decision-making.展开更多
The contemporary era is characterized by rapid technological advancements,particularly in the fields of communication and multimedia.Digital media has significantly influenced the daily lives of individuals of all age...The contemporary era is characterized by rapid technological advancements,particularly in the fields of communication and multimedia.Digital media has significantly influenced the daily lives of individuals of all ages.One of the emerging domains in digital media is the creation of cartoons and animated videos.The accessibility of the internet has led to a surge in the consumption of cartoons among young children,presenting challenges in monitoring and controlling the content they view.The prevalence of cartoon videos containing potentially violent scenes has raised concerns regarding their impact,especially on young and impressionableminds.This article contributes to the growing concerns about the impact of animated media on children’s mental health and offers solutions to help mitigate these effects.To address this issue,an intelligent,multi-CNN fusion framework is proposed for detecting and predicting violent content in upcoming frames of animated videos.The framework integrates probabilistic and deep learning methodologies by leveraging a combination of visual and temporal features for violence prediction in future scenes.Two specific convolutional neural network classifiers i.e.,VGG16 and ResNet18 are employed to classify scenes from animated content as violent or non-violent.To enhance decision robustness,this study introduces a fusion strategy based on weighted averaging,combining the outputs of both Convolutional Neural Networks(CNNs)into a single decision stream.The resulting classifications are subsequently fed into Naive Bayes classifier,which analyzes sequential patterns to forecast violence in future scenes.The experimental findings demonstrate that the proposed framework achieved predictive accuracy of 92.84%,highlighting its effectiveness for intelligent content moderation.These results underscore the potential of intelligent data fusion techniques in enhancing the reliability and robustness of automated violence detection systems in animated content.This framework offers a promising solution for safeguarding young audiences by enabling proactive and accurate moderation of animated videos.展开更多
An axial piston pump is a key component that plays the role of the "heart" in hydraulic systems. The pump failure will lead to an unexpected breakdown of the entire hydraulic system or even economic loss and...An axial piston pump is a key component that plays the role of the "heart" in hydraulic systems. The pump failure will lead to an unexpected breakdown of the entire hydraulic system or even economic loss and catastrophic safety consequences. Several vibration-based machine learning methods have been developed to detect and diagnose faults of axial piston pumps. However,most of these intelligent diagnosis methods use single-sensor vibration data to monitor the pump health states. Additionally, the diagnostic accuracy is unacceptable in most situations due to the complex pump structure and limited sensor information.Therefore, this study proposes a multi-sensor fusion method to improve the fault diagnosis performance of axial piston pumps.The convolutional neural network receives three channels of vibration data and makes the final diagnosis through information fusion at the decision level. The proposed decision fusion method is evaluated on the classification task of leakage levels of an actual axial piston pump. The experimental results show that the proposed method improves the classification accuracy by adjusting the probability distribution of classification according to the learned weight matrix.展开更多
Focusing on the problem of goal event detection in soccer videos,a novel method based on Hidden Markov Model(HMM) and the semantic rule is proposed.Firstly,a HMM for a goal event is constructed.Then a Normalized Seman...Focusing on the problem of goal event detection in soccer videos,a novel method based on Hidden Markov Model(HMM) and the semantic rule is proposed.Firstly,a HMM for a goal event is constructed.Then a Normalized Semantic Weighted Sum(NSWS) rule is established by defining a new feature of shots,semantic observation weight.The test video is detected based on the HMM and the NSWS rule,respectively.Finally,a fusion scheme based on logic distance is proposed and the detection results of the HMM and the NSWS rule are fused by optimal weights in the decision level,obtaining the final result.Experimental results indicate that the proposed method achieves 96.43% precision and 100% recall,which shows the effectiveness of this letter.展开更多
Background:Salvia miltiorrhiza Bunge,commonly known as“Danshen”in China due to the distinctive red color of its roots,is one of the most widely used traditional Chinese medicines.It is cultivated in various regions ...Background:Salvia miltiorrhiza Bunge,commonly known as“Danshen”in China due to the distinctive red color of its roots,is one of the most widely used traditional Chinese medicines.It is cultivated in various regions across China,and environmental differences among these regions can affect the secondary metabolites of plants,thereby influencing the quality of S.miltiorrhiza.In recent years,increasing demand for S.miltiorrhiza has exacerbated the problem of origin fraud.Therefore,ensuring the authenticity of its geographical origin is crucial for the sustainable development of the industry.Objective:The red coloration of S.miltiorrhiza is closely associated with the content of its primary active compounds,particularly tanshinones.Therefore,both its internal chemical composition and external color characteristics serve as key indicators for quality assessment.This study utilized hyperspectral imaging technology to evaluate its potential in classifying the geographical origin of S.miltiorrhiza.Methods:Spectral data reflecting the internal chemical properties of S.miltiorrhiza were integrated with color information representing its external features through 3 levels of data fusion.These fused datasets were then combined with deep learning algorithms to achieve accurate origin classification.Results:The results demonstrated that the Transformer model combined with soft-voting decision-level fusion achieved the highest classification accuracy of 98.72%by integrating image color and short-wave infrared spectral data.Conclusion:This study demonstrates that integrating hyperspectral imaging spectral data with color information provides a reliable and innovative approach for verifying the authenticity and traceability of S.miltiorrhiza.展开更多
Most present research into facial expression recognition focuses on the visible spectrum, which is sen- sitive to illumination change. In this paper, we focus on in- tegrating thermal infrared data with visible spectr...Most present research into facial expression recognition focuses on the visible spectrum, which is sen- sitive to illumination change. In this paper, we focus on in- tegrating thermal infrared data with visible spectrum images for spontaneous facial expression recognition. First, the ac- tive appearance model AAM parameters and three defined head motion features are extracted from visible spectrum im- ages, and several thermal statistical features are extracted from infrared （IR） images. Second, feature selection is per- formed using the F-test statistic. Third, Bayesian networks BNs and support vector machines SVMs are proposed for both decision-level and feature-level fusion. Experiments on the natural visible and infrared facial expression （NVIE） spontaneous database show the effectiveness of the proposed methods, and demonstrate thermal 1R images＇ supplementary role for visible facial expression recognition.展开更多
As a natural disaster,extreme precipitation is among the most destructive and influential,but predicting its occurrence and evolution accurately is very challenging because of its rarity and uniqueness.Taking the exam...As a natural disaster,extreme precipitation is among the most destructive and influential,but predicting its occurrence and evolution accurately is very challenging because of its rarity and uniqueness.Taking the example of the“21·7”extreme precipitation event(17–21 July 2021)in Henan Province,this study explores the potential of using physics-guided machine learning to improve the accuracy of forecasting the intensity and location of extreme precipitation.Three physics-guided ways of embedding physical features,fusing physical model forecasts and revised loss function are used,i.e.,(1)analyzing the anomalous circulation and thermodynamical factors,(2)analyzing the multi-model forecast bias and the associated underlying reasons for it,and(3)using professional forecasting knowledge to design the loss function,and the corresponding results are used as input for machine learning to improve the forecasting accuracy.The results indicate that by learning the relationship between anomalous physical features and heavy precipitation,the forecasting of precipitation intensity is improved significantly,but the location is rarely adjusted and more false alarms appear.Possible reasons for this are as follows.The anomalous features used here mainly contain information about large-scale systems and factors which are consistent with the model precipitation deviation;moreover,the samples of extreme precipitation are sparse and so the algorithm used here is simple.However,by combining“good and different”multi models with machine learning,the advantages of each model are extracted and then the location of the precipitation center in the forecast is improved significantly.Therefore,by combining the appropriate anomalous features with multi-model fusion,an integrated improvement of the forecast of the rainfall intensity and location is achieved.Overall,this study is a novel exploration to improve the refined forecasting of heavy precipitation with extreme intensity and high variability,and provides a reference for the deep fusion of physics and artificial intelligence methods to improve intense rain forecast.展开更多
基金Supported by Traditional Chinese Medicine Standardization Project of National Administration of Traditional Chinese Medicine:Research on the Physical Characteristics and Pre-disease Health Management of the Elderly in Hubei Province(No.GZY-FJS-2022-046)。
文摘OBJECTIVE:To develop an automated system for identifying and classifying constitution types in Traditional Chinese Medicine(TCM)by leveraging multi-model fusion algorithms.METHODS:A condensed version of a physical information collection form was designed to facilitate efficient data acquisition.The collected data were analyzed using a multi-model fusion approach,which integrated several machine learning techniques.These included support vector machines,Naive Bayes,decision trees,random forests,logistic regression,multilayer perceptrons,K-nearest neighbors,gradient boosting,adaptive ensemble learning,and recurrent neural networks.A soft voting strategy was used to combine the predictive outputs of each model,enabling the selection of the most effective model combination.RESULTS:The classification models demonstrated consistent and robust performance across most TCM constitution types when enhanced by the multi-model fusion strategy.In particular,high levels of accuracy,precision,recall,and F1-score were achieved for constitution types such as Yang deficiency,Qi deficiency,and Qi stagnation.However,the classification performance for the Yin deficiency constitution was relatively lower,indicating the need for further refinement and optimization in future research.CONCLUSION:This study introduces a novel,automated method for classifying TCM constitution types through the application of multi-model fusion algorithms.The approach simplifies the complex task of constitution identification while offering a practical and theoretical framework for the intelligent diagnosis of TCM body types.The findings have the potential to enhance personalized health management and support clinical decision-making in TCM diagnosis and treatment.
文摘[Objectives]This study was conducted to achieve rapid and accurate detection of protein content in rice with a particle size of 1.0 mm.[Methods]A multi-model fusion strategy was proposed on the basis of Stacking ensemble learning.A base learner pool was constructed,containing Partial Least Squares(PLS),Support Vector Machine(SVM),Deep Extreme Learning Machine(DELM),Random Forest(RF),Gradient Boosting Decision Tree(GBDT),and Multilayer Perceptron(MLP).PLS,DELM,and Linear Regression(LR)were used as meta-learner candidates.Employing integer coding technology,systematic dynamic combinations of base learners and meta-learners were generated,resulting in a total of 40 non-repetitive fusion models.The optimal combination was selected through a comprehensive evaluation based on multiple assessment indicators.[Results]The combination"PLS-DELM-MLP-LR"(code 1367)achieved coefficients of determination of 0.9732 and 0.9780 on the validation set and independent test set,respectively,with relative root mean square errors of 2.35%and 2.36%,and residual predictive deviations of 6.1075 and 6.7479,respectively.[Conclusions]The Stacking fusion model significantly enhances the predictive accuracy and robustness of spectral quantitative analysis,providing an efficient and feasible solution for modeling complex agricultural product spectral data.
基金supported by the National Natural Science Foundation of China (grant numbers 42293351, and U2468221)。
文摘This paper addresses the accuracy and timeliness limitations of traditional comprehensive prediction methods by proposing an approach of decision-level fusion of multisource data.A risk prediction indicator system was established for water and mud inrush in tunnels by analyzing advanced prediction data for specifi c tunnel segments.Additionally,the indicator weights were determined using the analytic hierarchy process combined with the Huber weighting method.Subsequently,a multisource data decision-layer fusion algorithm was utilized to generate fused imaging results for tunnel water and mud inrush risk predictions.Meanwhile,risk analysis was performed for different tunnel sections to achieve spatial and temporal complementarity within the indicator system and optimize redundant information.Finally,model feasibility was validated using the CZ Project Sejila Mountain Tunnel segment as a case study,yielding favorable risk prediction results and enabling effi cient information fusion and support for construction decision-making.
基金funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R138),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘The contemporary era is characterized by rapid technological advancements,particularly in the fields of communication and multimedia.Digital media has significantly influenced the daily lives of individuals of all ages.One of the emerging domains in digital media is the creation of cartoons and animated videos.The accessibility of the internet has led to a surge in the consumption of cartoons among young children,presenting challenges in monitoring and controlling the content they view.The prevalence of cartoon videos containing potentially violent scenes has raised concerns regarding their impact,especially on young and impressionableminds.This article contributes to the growing concerns about the impact of animated media on children’s mental health and offers solutions to help mitigate these effects.To address this issue,an intelligent,multi-CNN fusion framework is proposed for detecting and predicting violent content in upcoming frames of animated videos.The framework integrates probabilistic and deep learning methodologies by leveraging a combination of visual and temporal features for violence prediction in future scenes.Two specific convolutional neural network classifiers i.e.,VGG16 and ResNet18 are employed to classify scenes from animated content as violent or non-violent.To enhance decision robustness,this study introduces a fusion strategy based on weighted averaging,combining the outputs of both Convolutional Neural Networks(CNNs)into a single decision stream.The resulting classifications are subsequently fed into Naive Bayes classifier,which analyzes sequential patterns to forecast violence in future scenes.The experimental findings demonstrate that the proposed framework achieved predictive accuracy of 92.84%,highlighting its effectiveness for intelligent content moderation.These results underscore the potential of intelligent data fusion techniques in enhancing the reliability and robustness of automated violence detection systems in animated content.This framework offers a promising solution for safeguarding young audiences by enabling proactive and accurate moderation of animated videos.
基金supported by the National Key R&D Program of China(Grant No.2020YFB2007202)the National Natural Science Foundation of China(Grant No.52005323)+1 种基金the National Postdoctoral Program for Innovative Talents(Grant No.BX20200210)the China Postdoctoral Science Foundation(Grant No.2019M660086)。
文摘An axial piston pump is a key component that plays the role of the "heart" in hydraulic systems. The pump failure will lead to an unexpected breakdown of the entire hydraulic system or even economic loss and catastrophic safety consequences. Several vibration-based machine learning methods have been developed to detect and diagnose faults of axial piston pumps. However,most of these intelligent diagnosis methods use single-sensor vibration data to monitor the pump health states. Additionally, the diagnostic accuracy is unacceptable in most situations due to the complex pump structure and limited sensor information.Therefore, this study proposes a multi-sensor fusion method to improve the fault diagnosis performance of axial piston pumps.The convolutional neural network receives three channels of vibration data and makes the final diagnosis through information fusion at the decision level. The proposed decision fusion method is evaluated on the classification task of leakage levels of an actual axial piston pump. The experimental results show that the proposed method improves the classification accuracy by adjusting the probability distribution of classification according to the learned weight matrix.
基金Supported by the National Natural Science Foundation of China (No. 61072110)the Industrial Tackling Project of Shaanxi Province (2010K06-20)the Natural Science Foundation of Shaanxi Province (SJ08F15)
文摘Focusing on the problem of goal event detection in soccer videos,a novel method based on Hidden Markov Model(HMM) and the semantic rule is proposed.Firstly,a HMM for a goal event is constructed.Then a Normalized Semantic Weighted Sum(NSWS) rule is established by defining a new feature of shots,semantic observation weight.The test video is detected based on the HMM and the NSWS rule,respectively.Finally,a fusion scheme based on logic distance is proposed and the detection results of the HMM and the NSWS rule are fused by optimal weights in the decision level,obtaining the final result.Experimental results indicate that the proposed method achieves 96.43% precision and 100% recall,which shows the effectiveness of this letter.
基金the National Key R&D Program of China(2024YFC3506800,2024YFC3506805)the Scientific and Technological Innovation Project of the China Academy of Chinese Medical Sciences(CI2023E002,CI2021B009)+2 种基金the Quality and Technical Service Platform for the Traditional Chinese Medicine Whole Industry Chain(2022-230-221)the China Agricultural Research System of MOF and MARA(CARS-21)the Fundamental Research Funds for the Central Public Welfare Research Institutes(ZZ16-YQ-040,ZZXT2023012)。
文摘Background:Salvia miltiorrhiza Bunge,commonly known as“Danshen”in China due to the distinctive red color of its roots,is one of the most widely used traditional Chinese medicines.It is cultivated in various regions across China,and environmental differences among these regions can affect the secondary metabolites of plants,thereby influencing the quality of S.miltiorrhiza.In recent years,increasing demand for S.miltiorrhiza has exacerbated the problem of origin fraud.Therefore,ensuring the authenticity of its geographical origin is crucial for the sustainable development of the industry.Objective:The red coloration of S.miltiorrhiza is closely associated with the content of its primary active compounds,particularly tanshinones.Therefore,both its internal chemical composition and external color characteristics serve as key indicators for quality assessment.This study utilized hyperspectral imaging technology to evaluate its potential in classifying the geographical origin of S.miltiorrhiza.Methods:Spectral data reflecting the internal chemical properties of S.miltiorrhiza were integrated with color information representing its external features through 3 levels of data fusion.These fused datasets were then combined with deep learning algorithms to achieve accurate origin classification.Results:The results demonstrated that the Transformer model combined with soft-voting decision-level fusion achieved the highest classification accuracy of 98.72%by integrating image color and short-wave infrared spectral data.Conclusion:This study demonstrates that integrating hyperspectral imaging spectral data with color information provides a reliable and innovative approach for verifying the authenticity and traceability of S.miltiorrhiza.
文摘Most present research into facial expression recognition focuses on the visible spectrum, which is sen- sitive to illumination change. In this paper, we focus on in- tegrating thermal infrared data with visible spectrum images for spontaneous facial expression recognition. First, the ac- tive appearance model AAM parameters and three defined head motion features are extracted from visible spectrum im- ages, and several thermal statistical features are extracted from infrared （IR） images. Second, feature selection is per- formed using the F-test statistic. Third, Bayesian networks BNs and support vector machines SVMs are proposed for both decision-level and feature-level fusion. Experiments on the natural visible and infrared facial expression （NVIE） spontaneous database show the effectiveness of the proposed methods, and demonstrate thermal 1R images＇ supplementary role for visible facial expression recognition.
基金supported by the National Key R&D Project(Grant No.2021YFC3000903)the National Natural Science Foundation of China(Grant Nos.42275013,42030611,42075002)+2 种基金the CMA Innovation Foundation(Grant No.CXFZ2023J001)the Open Grants of the State Key Laboratory of Severe Weather(Grant No.2023LASW-B05)the Key Foundation of Zhejiang Provincial Department of Science and Technology(Grant No.2022C03150)。
文摘As a natural disaster,extreme precipitation is among the most destructive and influential,but predicting its occurrence and evolution accurately is very challenging because of its rarity and uniqueness.Taking the example of the“21·7”extreme precipitation event(17–21 July 2021)in Henan Province,this study explores the potential of using physics-guided machine learning to improve the accuracy of forecasting the intensity and location of extreme precipitation.Three physics-guided ways of embedding physical features,fusing physical model forecasts and revised loss function are used,i.e.,(1)analyzing the anomalous circulation and thermodynamical factors,(2)analyzing the multi-model forecast bias and the associated underlying reasons for it,and(3)using professional forecasting knowledge to design the loss function,and the corresponding results are used as input for machine learning to improve the forecasting accuracy.The results indicate that by learning the relationship between anomalous physical features and heavy precipitation,the forecasting of precipitation intensity is improved significantly,but the location is rarely adjusted and more false alarms appear.Possible reasons for this are as follows.The anomalous features used here mainly contain information about large-scale systems and factors which are consistent with the model precipitation deviation;moreover,the samples of extreme precipitation are sparse and so the algorithm used here is simple.However,by combining“good and different”multi models with machine learning,the advantages of each model are extracted and then the location of the precipitation center in the forecast is improved significantly.Therefore,by combining the appropriate anomalous features with multi-model fusion,an integrated improvement of the forecast of the rainfall intensity and location is achieved.Overall,this study is a novel exploration to improve the refined forecasting of heavy precipitation with extreme intensity and high variability,and provides a reference for the deep fusion of physics and artificial intelligence methods to improve intense rain forecast.
