In view of the problem that it＇s difficult to accurately grasp the influence range and transmission path of the vehicle top design requirements on the underlying design parameters. Applying directed-weighted complex ...In view of the problem that it＇s difficult to accurately grasp the influence range and transmission path of the vehicle top design requirements on the underlying design parameters. Applying directed-weighted complex network to product parameter model is an important method that can clarify the relationships between product parameters and establish the top-down design of a product. The relationships of the product parameters of each node are calculated via a simple path searching algorithm, and the main design parameters are extracted by analysis and comparison. A uniform definition of the index formula for out-in degree can be provided based on the analysis of out- in-degree width and depth and control strength of train carriage body parameters. Vehicle gauge, axle load, crosswind and other parameters with higher values of the out-degree index are the most important boundary condi- tions; the most considerable performance indices are the parameters that have higher values of the out-in-degree index including torsional stiffness, maximum testing speed, service life of the vehicle, and so on; the main design parameters contain train carriage body weight, train weight per extended metre, train height and other parameters with higher values of the in-degree index. The network not only provides theoretical guidance for exploring the relationship of design parameters, but also further enriches the appli- cation of forward design method to high-speed trains.展开更多
Recognition of human activity based on convolutional neural network(CNN)has received the interest of researchers in recent years due to its significant improvement in accuracy.A large number of algorithms based on the...Recognition of human activity based on convolutional neural network(CNN)has received the interest of researchers in recent years due to its significant improvement in accuracy.A large number of algorithms based on the deep learning approach have been proposed for activity recognition purpose.However,with the increasing advancements in technologies having limited computational resources,it needs to design an efficient deep learning-based approaches with improved utilization of computational resources.This paper presents a simple and efficient 2-dimensional CNN(2-D CNN)architecture with very small-size convolutional kernel for human activity recognition.The merit of the proposed CNN architecture over standard deep learning architectures is fewer trainable parameters and lesser memory requirement which enables it to train the proposed CNN architecture on low GPU memory-based devices and also works well with smaller as well as larger size datasets.The proposed approach consists of mainly four stages:namely(1)creation of dataset and data augmentation,(2)designing 2-D CNN architecture,(3)the proposed 2-D CNN architecture trained from scratch up to optimum stage,and(4)evaluation of the trained 2-D CNN architecture.To illustrate the effectiveness of the proposed architecture several extensive experiments are conducted on three publicly available datasets,namely IXMAS,YouTube,and UCF101 dataset.The results of the proposed method and its comparison with other state-of-the-art methods demonstrate the usefulness of the proposed method.展开更多
Lung cancer is one of the malignancies with the highest incidence and mortality rates worldwide.Accurate detection of early lung nodules is crucial for improving patient survival.However,manual reading of medical imag...Lung cancer is one of the malignancies with the highest incidence and mortality rates worldwide.Accurate detection of early lung nodules is crucial for improving patient survival.However,manual reading of medical images carries risks of missed or incorrect diagnoses,highlighting the urgent need for efficient and precise automated detection methods.This study constructs and optimizes a lung nodule detection model based on You Only Look Once version 8(YOLOv8).Model adjustments include tuning learning rate strategies,weight decay,and color augmentation parameters.Experiments were conducted on the standard LIDC-IDRI dataset.Model performance was evaluated using recall,Mean Average Precision(mAP),and the confusion matrix.The results indicate a recall rate of 0.87 and an mAP of 0.776,with a correct lung nodule recognition rate of 0.76 in the confusion matrix.The study demonstrates that rational parameter optimization can effectively reduce missed detections and misjudgments while significantly improving detection accuracy and classification stability.This provides a feasible technical pathway for computer-aided early lung cancer diagnosis.Future work may further enhance the model’s application value in intelligent medical imaging analysis by integrating data augmentation with parameter optimization.展开更多
Artificial intelligence(AI)is profoundly reshaping the practical logic and theoretical boundaries of bel canto education.This study takes“technical path-aesthetic challenge-paradigm reconstruction”as the analytical ...Artificial intelligence(AI)is profoundly reshaping the practical logic and theoretical boundaries of bel canto education.This study takes“technical path-aesthetic challenge-paradigm reconstruction”as the analytical framework to systematically explore the innovative mechanism and practical path of AI-empowered vocal education.At the technical level,an acoustic parameter modeling system is constructed based on deep learning.Through MEL spectrum analysis,forresonance tracking and breath dynamics modeling,the traditional singing assessment that relies on auditory experience is transformed into quantifiable and interpretable acoustic indicators(such as harmonic energy ratio,vowel resonance stability,etc.),achieving precise diagnosis and real-time intervention of vocal defects.Experimental data shows that this system reduces the learners’pitch error rate by 42%and increases their breath control efficiency by 35%.In the dimension of personalized training,an adaptive teaching engine based on reinforcement learning was developed to dynamically generate a three-dimensional matching scheme of“voice features-track difficulty-training intensity”,significantly reducing the time for beginners to master core vocal techniques(by an average of 28%).However,the conflict between AI quantitative indicators and the traditional aesthetic standards of vocal music has become prominent:the algorithm’s preference for standardized formuster distribution may suppress the singer’s unique timbre personality and artistic expression tension.To this end,a“dual-track evaluation model”is proposed:taking acoustic parameters as the technical benchmark and emotional expression and artistic appeal as the aesthetic benchmark,and achieving a dynamic balance between the two through expert annotation and group consensus algorithms.The research further reconstructs the teaching paradigm,advocating that AI tools be positioned as“aesthetic collaborators”rather than substitutes,and builds a new type of teacher-student relationship of“human-machine co-teaching-co-evaluation-co-creation”.This research provides a solution that is both technically feasible and aesthetically reasonable for the digital transformation of bel canto education,revealing the underlying logic of the deep integration of art and technology in the AI era.展开更多
The limited understanding of the microstructure and dynamic evolution associated with the nonstoichiometric characteristics of wustite has constrained the comprehension of iron oxide properties,diffusion,and phase tra...The limited understanding of the microstructure and dynamic evolution associated with the nonstoichiometric characteristics of wustite has constrained the comprehension of iron oxide properties,diffusion,and phase transformation behaviors.This study employs deep learning methods to train interatomic potential parameters for the Fe–O system,achieving precise atomic-scale simulations of the wustite phase structure and internal lattice defects.This approach addresses the shortcomings of large-scale molecular dynamics simulations in accurately describing the solid-phase structure of the Fe–O system.Utilizing these potential parameters,this research is the first to reveal the complex mechanisms underlying the non-stoichiometric nature of wustite(Fe_(1−x)O).The study found that cation vacancy defects in wustite tend to aggregate,forming stable cluster structures.It also elucidated the formation mechanisms of interstitial iron atoms and typical defect clusters in wustite,establishing the formation preference for Koch–Cohen defect clusters.These potential parameters and research methods can be further applied in future studies on iron oxide reduction,phase transformation mechanisms,and related material development,thereby advancing fundamental research in metallurgy and related industries.展开更多
基金Supported by National Natural Science Foundation of China(Grant Nos51275432,51505390)Sichuan Provincial Application Foundation Projects of China(Grant No.2016JY0098)Independent Research Project of TPL(Grant No.TPL1501)
文摘In view of the problem that it＇s difficult to accurately grasp the influence range and transmission path of the vehicle top design requirements on the underlying design parameters. Applying directed-weighted complex network to product parameter model is an important method that can clarify the relationships between product parameters and establish the top-down design of a product. The relationships of the product parameters of each node are calculated via a simple path searching algorithm, and the main design parameters are extracted by analysis and comparison. A uniform definition of the index formula for out-in degree can be provided based on the analysis of out- in-degree width and depth and control strength of train carriage body parameters. Vehicle gauge, axle load, crosswind and other parameters with higher values of the out-degree index are the most important boundary condi- tions; the most considerable performance indices are the parameters that have higher values of the out-in-degree index including torsional stiffness, maximum testing speed, service life of the vehicle, and so on; the main design parameters contain train carriage body weight, train weight per extended metre, train height and other parameters with higher values of the in-degree index. The network not only provides theoretical guidance for exploring the relationship of design parameters, but also further enriches the appli- cation of forward design method to high-speed trains.
文摘Recognition of human activity based on convolutional neural network(CNN)has received the interest of researchers in recent years due to its significant improvement in accuracy.A large number of algorithms based on the deep learning approach have been proposed for activity recognition purpose.However,with the increasing advancements in technologies having limited computational resources,it needs to design an efficient deep learning-based approaches with improved utilization of computational resources.This paper presents a simple and efficient 2-dimensional CNN(2-D CNN)architecture with very small-size convolutional kernel for human activity recognition.The merit of the proposed CNN architecture over standard deep learning architectures is fewer trainable parameters and lesser memory requirement which enables it to train the proposed CNN architecture on low GPU memory-based devices and also works well with smaller as well as larger size datasets.The proposed approach consists of mainly four stages:namely(1)creation of dataset and data augmentation,(2)designing 2-D CNN architecture,(3)the proposed 2-D CNN architecture trained from scratch up to optimum stage,and(4)evaluation of the trained 2-D CNN architecture.To illustrate the effectiveness of the proposed architecture several extensive experiments are conducted on three publicly available datasets,namely IXMAS,YouTube,and UCF101 dataset.The results of the proposed method and its comparison with other state-of-the-art methods demonstrate the usefulness of the proposed method.
基金The Undergraduate Innovation and Entrepreneurship Project-Intelligent Medical Assistance System for Lung Nodule Detection Based on Deep Learning(Project No.S202510656067)The Special Fund for Basic Scientific Research of Central Universities at Southwest Minzu University(Approval No.ZYN2024069)。
文摘Lung cancer is one of the malignancies with the highest incidence and mortality rates worldwide.Accurate detection of early lung nodules is crucial for improving patient survival.However,manual reading of medical images carries risks of missed or incorrect diagnoses,highlighting the urgent need for efficient and precise automated detection methods.This study constructs and optimizes a lung nodule detection model based on You Only Look Once version 8(YOLOv8).Model adjustments include tuning learning rate strategies,weight decay,and color augmentation parameters.Experiments were conducted on the standard LIDC-IDRI dataset.Model performance was evaluated using recall,Mean Average Precision(mAP),and the confusion matrix.The results indicate a recall rate of 0.87 and an mAP of 0.776,with a correct lung nodule recognition rate of 0.76 in the confusion matrix.The study demonstrates that rational parameter optimization can effectively reduce missed detections and misjudgments while significantly improving detection accuracy and classification stability.This provides a feasible technical pathway for computer-aided early lung cancer diagnosis.Future work may further enhance the model’s application value in intelligent medical imaging analysis by integrating data augmentation with parameter optimization.
文摘Artificial intelligence(AI)is profoundly reshaping the practical logic and theoretical boundaries of bel canto education.This study takes“technical path-aesthetic challenge-paradigm reconstruction”as the analytical framework to systematically explore the innovative mechanism and practical path of AI-empowered vocal education.At the technical level,an acoustic parameter modeling system is constructed based on deep learning.Through MEL spectrum analysis,forresonance tracking and breath dynamics modeling,the traditional singing assessment that relies on auditory experience is transformed into quantifiable and interpretable acoustic indicators(such as harmonic energy ratio,vowel resonance stability,etc.),achieving precise diagnosis and real-time intervention of vocal defects.Experimental data shows that this system reduces the learners’pitch error rate by 42%and increases their breath control efficiency by 35%.In the dimension of personalized training,an adaptive teaching engine based on reinforcement learning was developed to dynamically generate a three-dimensional matching scheme of“voice features-track difficulty-training intensity”,significantly reducing the time for beginners to master core vocal techniques(by an average of 28%).However,the conflict between AI quantitative indicators and the traditional aesthetic standards of vocal music has become prominent:the algorithm’s preference for standardized formuster distribution may suppress the singer’s unique timbre personality and artistic expression tension.To this end,a“dual-track evaluation model”is proposed:taking acoustic parameters as the technical benchmark and emotional expression and artistic appeal as the aesthetic benchmark,and achieving a dynamic balance between the two through expert annotation and group consensus algorithms.The research further reconstructs the teaching paradigm,advocating that AI tools be positioned as“aesthetic collaborators”rather than substitutes,and builds a new type of teacher-student relationship of“human-machine co-teaching-co-evaluation-co-creation”.This research provides a solution that is both technically feasible and aesthetically reasonable for the digital transformation of bel canto education,revealing the underlying logic of the deep integration of art and technology in the AI era.
基金the support of the Young Elite Scientist Sponsorship Program by CAST(YESS20210090)Beijing Natural Science Foundation(J210017)China Baowu Low Carbon Metallurgy Innovation Foundation-BWLCF202119.
文摘The limited understanding of the microstructure and dynamic evolution associated with the nonstoichiometric characteristics of wustite has constrained the comprehension of iron oxide properties,diffusion,and phase transformation behaviors.This study employs deep learning methods to train interatomic potential parameters for the Fe–O system,achieving precise atomic-scale simulations of the wustite phase structure and internal lattice defects.This approach addresses the shortcomings of large-scale molecular dynamics simulations in accurately describing the solid-phase structure of the Fe–O system.Utilizing these potential parameters,this research is the first to reveal the complex mechanisms underlying the non-stoichiometric nature of wustite(Fe_(1−x)O).The study found that cation vacancy defects in wustite tend to aggregate,forming stable cluster structures.It also elucidated the formation mechanisms of interstitial iron atoms and typical defect clusters in wustite,establishing the formation preference for Koch–Cohen defect clusters.These potential parameters and research methods can be further applied in future studies on iron oxide reduction,phase transformation mechanisms,and related material development,thereby advancing fundamental research in metallurgy and related industries.
