Consulting concepts of business process reengineering of modem enterprises and according to networked industry characteristics of railway, this article analyzes restructuring experience modes of foreign railway, probe...Consulting concepts of business process reengineering of modem enterprises and according to networked industry characteristics of railway, this article analyzes restructuring experience modes of foreign railway, probes into the trend of China railway restructure, and makes it clear that railway should set up modem enterprise systems by focusing on transport as major trade. As a result, the railway informalization system shall be guided by marketing, aiming at improving the operation efficiency and management benefits, having priority in passenger and goods marketing system development. Central control and safeguarding systems should be emphasized in construction and railway business management systems shall be broueht about.展开更多
Border crossing is a metaphoric term in sociolinguistics referring to a blurring of English use across different domains of social discourses.This article shows how border crossing in English job-seeking discourse is ...Border crossing is a metaphoric term in sociolinguistics referring to a blurring of English use across different domains of social discourses.This article shows how border crossing in English job-seeking discourse is manifested in a real-life English creative résumé,and how this phenomenon brings about two related consequences,informalization and marketization,through critically analyzing the informal and promotional features of the résumé.It then reaches the conclusion that such border crossing phenomenon reflects a voice by the young in reconstructing their self-identities in job-seeking discourse in post-industrialized,post-modern western societies.展开更多
Since the 1980s,along with the transformation to the market economy,the most tangible feature in the legal development in China is toward the legalization:China has made more laws,tried more civil and criminal cases,h...Since the 1980s,along with the transformation to the market economy,the most tangible feature in the legal development in China is toward the legalization:China has made more laws,tried more civil and criminal cases,had more lawyers,judges and prosecutors,set up more legal educational institutes.Not only on the book but also in action,Chinese people more and more like to choose the legal way,the people’s courts.展开更多
Traffic sign detection is a critical component of driving systems.Single-stage network-based traffic sign detection algorithms,renowned for their fast detection speeds and high accuracy,have become the dominant approa...Traffic sign detection is a critical component of driving systems.Single-stage network-based traffic sign detection algorithms,renowned for their fast detection speeds and high accuracy,have become the dominant approach in current practices.However,in complex and dynamic traffic scenes,particularly with smaller traffic sign objects,challenges such as missed and false detections can lead to reduced overall detection accuracy.To address this issue,this paper proposes a detection algorithm that integrates edge and shape information.Recognizing that traffic signs have specific shapes and distinct edge contours,this paper introduces an edge feature extraction branch within the backbone network,enabling adaptive fusion with features of the same hierarchical level.Additionally,a shape prior convolution module is designed to replaces the first two convolutional modules of the backbone network,aimed at enhancing the model's perception ability for specific shape objects and reducing its sensitivity to background noise.The algorithm was evaluated on the CCTSDB and TT100k datasets,and compared to YOLOv8s,the mAP50 values increased by 3.0%and 10.4%,respectively,demonstrating the effectiveness of the proposed method in improving the accuracy of traffic sign detection.展开更多
Experimental therapies targeting immune and stromal cells,such as mast cells,cancer-associated fibroblasts,dendritic cells,and tumor endothelial cells,in the treatment of gastrointestinal solid tumors pose new and com...Experimental therapies targeting immune and stromal cells,such as mast cells,cancer-associated fibroblasts,dendritic cells,and tumor endothelial cells,in the treatment of gastrointestinal solid tumors pose new and complex surgical and medico-legal challenges.These innovative treatments require that informed consent not be limited to simple acceptance of the medical procedure,but instead reflect a true relational and cognitive process grounded in understanding,free choice,and the ability to revoke consent at any time.In particular,it is essential that the patient understands the experimental nature of the therapy,its development stage,potential benefits and risks,as well as the implications for their health and personal dignity.In the case of stromal cell-based treatments,which may exert complex immunomodulatory effects or activate angiogenic pathways that are not yet fully understood,patients must be made fully aware that they are participating in a non-standardized therapy whose outcomes,whether beneficial or harmful,cannot yet be predicted with certainty.This requires particularly careful medical communication,using simple yet scientifically accurate explanations delivered in appropriate language,along with a final verification of the patient’s actual understanding.展开更多
Quantitative analysis of aluminum-silicon(Al-Si)alloy microstructure is crucial for evaluating and controlling alloy performance.Conventional analysis methods rely on manual segmentation,which is inefficient and subje...Quantitative analysis of aluminum-silicon(Al-Si)alloy microstructure is crucial for evaluating and controlling alloy performance.Conventional analysis methods rely on manual segmentation,which is inefficient and subjective,while fully supervised deep learning approaches require extensive and expensive pixel-level annotated data.Furthermore,existing semi-supervised methods still face challenges in handling the adhesion of adjacent primary silicon particles and effectively utilizing consistency in unlabeled data.To address these issues,this paper proposes a novel semi-supervised framework for Al-Si alloy microstructure image segmentation.First,we introduce a Rotational Uncertainty Correction Strategy(RUCS).This strategy employs multi-angle rotational perturbations andMonte Carlo sampling to assess prediction consistency,generating a pixel-wise confidence weight map.By integrating this map into the loss function,the model dynamically focuses on high-confidence regions,thereby improving generalization ability while reducing manual annotation pressure.Second,we design a Boundary EnhancementModule(BEM)to strengthen boundary feature extraction through erosion difference and multi-scale dilated convolutions.This module guides the model to focus on the boundary regions of adjacent particles,effectively resolving particle adhesion and improving segmentation accuracy.Systematic experiments were conducted on the Aluminum-Silicon Alloy Microstructure Dataset(ASAD).Results indicate that the proposed method performs exceptionally well with scarce labeled data.Specifically,using only 5%labeled data,our method improves the Jaccard index and Adjusted Rand Index(ARI)by 2.84 and 1.57 percentage points,respectively,and reduces the Variation of Information(VI)by 8.65 compared to stateof-the-art semi-supervised models,approaching the performance levels of 10%labeled data.These results demonstrate that the proposed method significantly enhances the accuracy and robustness of quantitative microstructure analysis while reducing annotation costs.展开更多
Submission.Papers appearing in the Journal comprise Editorials,Rapid Communications,Perspectives,Tutorials,Feature Articles,Reviews,Research Articles,which should contain original information,theoretical or experiment...Submission.Papers appearing in the Journal comprise Editorials,Rapid Communications,Perspectives,Tutorials,Feature Articles,Reviews,Research Articles,which should contain original information,theoretical or experimental,on any topics in the field of polymer science and polymer material science.Papers already published or scheduled to be published elsewhere should not be submitted and certainly will not be accepted.展开更多
文摘Consulting concepts of business process reengineering of modem enterprises and according to networked industry characteristics of railway, this article analyzes restructuring experience modes of foreign railway, probes into the trend of China railway restructure, and makes it clear that railway should set up modem enterprise systems by focusing on transport as major trade. As a result, the railway informalization system shall be guided by marketing, aiming at improving the operation efficiency and management benefits, having priority in passenger and goods marketing system development. Central control and safeguarding systems should be emphasized in construction and railway business management systems shall be broueht about.
文摘Border crossing is a metaphoric term in sociolinguistics referring to a blurring of English use across different domains of social discourses.This article shows how border crossing in English job-seeking discourse is manifested in a real-life English creative résumé,and how this phenomenon brings about two related consequences,informalization and marketization,through critically analyzing the informal and promotional features of the résumé.It then reaches the conclusion that such border crossing phenomenon reflects a voice by the young in reconstructing their self-identities in job-seeking discourse in post-industrialized,post-modern western societies.
文摘Since the 1980s,along with the transformation to the market economy,the most tangible feature in the legal development in China is toward the legalization:China has made more laws,tried more civil and criminal cases,had more lawyers,judges and prosecutors,set up more legal educational institutes.Not only on the book but also in action,Chinese people more and more like to choose the legal way,the people’s courts.
基金supported by the National Natural Science Foundation of China(Grant Nos.62572057,62272049,U24A20331)Beijing Natural Science Foundation(Grant Nos.4232026,4242020)Academic Research Projects of Beijing Union University(Grant No.ZK10202404).
文摘Traffic sign detection is a critical component of driving systems.Single-stage network-based traffic sign detection algorithms,renowned for their fast detection speeds and high accuracy,have become the dominant approach in current practices.However,in complex and dynamic traffic scenes,particularly with smaller traffic sign objects,challenges such as missed and false detections can lead to reduced overall detection accuracy.To address this issue,this paper proposes a detection algorithm that integrates edge and shape information.Recognizing that traffic signs have specific shapes and distinct edge contours,this paper introduces an edge feature extraction branch within the backbone network,enabling adaptive fusion with features of the same hierarchical level.Additionally,a shape prior convolution module is designed to replaces the first two convolutional modules of the backbone network,aimed at enhancing the model's perception ability for specific shape objects and reducing its sensitivity to background noise.The algorithm was evaluated on the CCTSDB and TT100k datasets,and compared to YOLOv8s,the mAP50 values increased by 3.0%and 10.4%,respectively,demonstrating the effectiveness of the proposed method in improving the accuracy of traffic sign detection.
文摘Experimental therapies targeting immune and stromal cells,such as mast cells,cancer-associated fibroblasts,dendritic cells,and tumor endothelial cells,in the treatment of gastrointestinal solid tumors pose new and complex surgical and medico-legal challenges.These innovative treatments require that informed consent not be limited to simple acceptance of the medical procedure,but instead reflect a true relational and cognitive process grounded in understanding,free choice,and the ability to revoke consent at any time.In particular,it is essential that the patient understands the experimental nature of the therapy,its development stage,potential benefits and risks,as well as the implications for their health and personal dignity.In the case of stromal cell-based treatments,which may exert complex immunomodulatory effects or activate angiogenic pathways that are not yet fully understood,patients must be made fully aware that they are participating in a non-standardized therapy whose outcomes,whether beneficial or harmful,cannot yet be predicted with certainty.This requires particularly careful medical communication,using simple yet scientifically accurate explanations delivered in appropriate language,along with a final verification of the patient’s actual understanding.
基金funded by the National Natural Science Foundation of China (52061020).
文摘Quantitative analysis of aluminum-silicon(Al-Si)alloy microstructure is crucial for evaluating and controlling alloy performance.Conventional analysis methods rely on manual segmentation,which is inefficient and subjective,while fully supervised deep learning approaches require extensive and expensive pixel-level annotated data.Furthermore,existing semi-supervised methods still face challenges in handling the adhesion of adjacent primary silicon particles and effectively utilizing consistency in unlabeled data.To address these issues,this paper proposes a novel semi-supervised framework for Al-Si alloy microstructure image segmentation.First,we introduce a Rotational Uncertainty Correction Strategy(RUCS).This strategy employs multi-angle rotational perturbations andMonte Carlo sampling to assess prediction consistency,generating a pixel-wise confidence weight map.By integrating this map into the loss function,the model dynamically focuses on high-confidence regions,thereby improving generalization ability while reducing manual annotation pressure.Second,we design a Boundary EnhancementModule(BEM)to strengthen boundary feature extraction through erosion difference and multi-scale dilated convolutions.This module guides the model to focus on the boundary regions of adjacent particles,effectively resolving particle adhesion and improving segmentation accuracy.Systematic experiments were conducted on the Aluminum-Silicon Alloy Microstructure Dataset(ASAD).Results indicate that the proposed method performs exceptionally well with scarce labeled data.Specifically,using only 5%labeled data,our method improves the Jaccard index and Adjusted Rand Index(ARI)by 2.84 and 1.57 percentage points,respectively,and reduces the Variation of Information(VI)by 8.65 compared to stateof-the-art semi-supervised models,approaching the performance levels of 10%labeled data.These results demonstrate that the proposed method significantly enhances the accuracy and robustness of quantitative microstructure analysis while reducing annotation costs.
文摘Submission.Papers appearing in the Journal comprise Editorials,Rapid Communications,Perspectives,Tutorials,Feature Articles,Reviews,Research Articles,which should contain original information,theoretical or experimental,on any topics in the field of polymer science and polymer material science.Papers already published or scheduled to be published elsewhere should not be submitted and certainly will not be accepted.
