Increasingly,attention is being directed towards time-dependent diffusion magnetic resonance imaging(TDDMRI),a method that reveals time-related changes in the diffusional behavior of water molecules in biological tiss...Increasingly,attention is being directed towards time-dependent diffusion magnetic resonance imaging(TDDMRI),a method that reveals time-related changes in the diffusional behavior of water molecules in biological tissues,thereby enabling us to probe related microstructure events.With ongoing improvements in hardware and advanced pulse sequences,significant progress has been made in applying TDDMRI to clinical research.The development of accurate mathematical models and computational methods has bolstered theoretical support for TDDMRI and elevated our understanding of molecular diffusion.In this review,we introduce the concept and basic physics of TDDMRI,and then focus on the measurement strategies and modeling approaches in short-and long-diffusion-time domains.Finally,we discuss the challenges in this field,including the requirement for efficient scanning and data processing technologies,the development of more precise models depicting time-dependent molecular diffusion,and critical clinical applications.展开更多
This study presents a deep learning framework for non-destructive evaluation of concrete compressive strength using high-resolution microstructural images.Unlike traditional destructive testing,this approach enables e...This study presents a deep learning framework for non-destructive evaluation of concrete compressive strength using high-resolution microstructural images.Unlike traditional destructive testing,this approach enables efficient largescale and continuous strength monitoring.The proposed model combines:(1)CAE for efficient feature extraction(achieving 80%dimensionality reduction without significant information loss);(2)Transformer-based self-attention mechanisms to dynamically weight critical image regions,enhancing interpretability;and(3)LSTM networks to capture temporal strength evolution during curing,improving forecasting accuracy by 15%.The framework is trained and tested on a hybrid dataset integrating UCI concrete strength data with high-resolution microstructural images.Nested cross-validation coupled with Bayesian optimization ensures robust performance evaluation and hyperparameter tuning.Comparative analyses demonstrate superior performance over baseline CNN and traditional ML models,with 20%reduction in MAE(3.7 MPa vs.4.6 MPa),18%lower RMSE(4.9 MPa vs.6.1 MPa),and 7%higher R2(0.87 vs.0.81).The model also reduces prediction time by approximately 20%.This scalable solution offers high accuracy,robustness,and generalizability for real-time concrete strength monitoring in infrastructure projects,advancing intelligent image-based non-destructive testing beyond conventional destructive methods.展开更多
The presence of nonmetallic inclusion particles which appear during steelmaking process is harmful to the properties of steels, which is mainly as a function of some aspects such as size, volume fraction, shape, and d...The presence of nonmetallic inclusion particles which appear during steelmaking process is harmful to the properties of steels, which is mainly as a function of some aspects such as size, volume fraction, shape, and distribution of these particles. The automatic image analysis technique is one of the most important tools for the quantitative determination of these parameters. The classical Student approach and the Extreme Values Method （EVM） were used for the inclusion size and shape determination and the evaluation of distance between the inclusion particles. The results thus obtained indicated that there were significant differences in the characteristics of the inclusion particles in the analyzed products. Both methods achieved results with some differences, indicating that EVM could be used as a faster and more reliable statistical methodology.展开更多
The main objective of this paper is to evaluate the effects of asphalt concrete types on the microstructural characteristics at high-temperature. Suspend-dense structure and Skeleton-dense structure were selected to i...The main objective of this paper is to evaluate the effects of asphalt concrete types on the microstructural characteristics at high-temperature. Suspend-dense structure and Skeleton-dense structure were selected to investigate the deformation of pavement at meso-scale. The internal microstructures of typical asphalt concretes, AC, SUP and SMA, were scanned by X-ray CT device, and microstructural changes before and after high-temperature damage were researched by digital image processing. Adaptive threshold segmentation algorithm（ATSA） based on image radius was developed and utilized to obtain the binary images of aggregates, air-voids and asphalt mastic. Then the shape and distribution of air-voids and aggregates were analyzed. The results show that the ATSA can distinguish the target and background effectively. Gradation and coarse aggregate size of asphalt mixtures have an obvious influence on the distribution of air-voids. The movements of aggregate particles are complex and aggregates with elliptic sharp show great rotation. The effect of gradation on microstructure during high-temperature damage promotes the research about the failure mechanism of asphalt concrete pavement.展开更多
基金supported by the Ministry of Science and Technology of the People’s Republic of China(No.2021ZD0200202)the National Natural Science Foundation of China(No.82122032)the Science and Technology Department of Zhejiang Province(Nos.202006140 and 2022C03057).
文摘Increasingly,attention is being directed towards time-dependent diffusion magnetic resonance imaging(TDDMRI),a method that reveals time-related changes in the diffusional behavior of water molecules in biological tissues,thereby enabling us to probe related microstructure events.With ongoing improvements in hardware and advanced pulse sequences,significant progress has been made in applying TDDMRI to clinical research.The development of accurate mathematical models and computational methods has bolstered theoretical support for TDDMRI and elevated our understanding of molecular diffusion.In this review,we introduce the concept and basic physics of TDDMRI,and then focus on the measurement strategies and modeling approaches in short-and long-diffusion-time domains.Finally,we discuss the challenges in this field,including the requirement for efficient scanning and data processing technologies,the development of more precise models depicting time-dependent molecular diffusion,and critical clinical applications.
文摘This study presents a deep learning framework for non-destructive evaluation of concrete compressive strength using high-resolution microstructural images.Unlike traditional destructive testing,this approach enables efficient largescale and continuous strength monitoring.The proposed model combines:(1)CAE for efficient feature extraction(achieving 80%dimensionality reduction without significant information loss);(2)Transformer-based self-attention mechanisms to dynamically weight critical image regions,enhancing interpretability;and(3)LSTM networks to capture temporal strength evolution during curing,improving forecasting accuracy by 15%.The framework is trained and tested on a hybrid dataset integrating UCI concrete strength data with high-resolution microstructural images.Nested cross-validation coupled with Bayesian optimization ensures robust performance evaluation and hyperparameter tuning.Comparative analyses demonstrate superior performance over baseline CNN and traditional ML models,with 20%reduction in MAE(3.7 MPa vs.4.6 MPa),18%lower RMSE(4.9 MPa vs.6.1 MPa),and 7%higher R2(0.87 vs.0.81).The model also reduces prediction time by approximately 20%.This scalable solution offers high accuracy,robustness,and generalizability for real-time concrete strength monitoring in infrastructure projects,advancing intelligent image-based non-destructive testing beyond conventional destructive methods.
文摘The presence of nonmetallic inclusion particles which appear during steelmaking process is harmful to the properties of steels, which is mainly as a function of some aspects such as size, volume fraction, shape, and distribution of these particles. The automatic image analysis technique is one of the most important tools for the quantitative determination of these parameters. The classical Student approach and the Extreme Values Method （EVM） were used for the inclusion size and shape determination and the evaluation of distance between the inclusion particles. The results thus obtained indicated that there were significant differences in the characteristics of the inclusion particles in the analyzed products. Both methods achieved results with some differences, indicating that EVM could be used as a faster and more reliable statistical methodology.
基金Funded by National Natural Science Foundation of China(No.51178114)the Fundamental Research Funds for the Central Universities(No.CXLX12_0117)the Scientific Research Foundation of Graduate School of Southeast University(No.YBJJ1318)
文摘The main objective of this paper is to evaluate the effects of asphalt concrete types on the microstructural characteristics at high-temperature. Suspend-dense structure and Skeleton-dense structure were selected to investigate the deformation of pavement at meso-scale. The internal microstructures of typical asphalt concretes, AC, SUP and SMA, were scanned by X-ray CT device, and microstructural changes before and after high-temperature damage were researched by digital image processing. Adaptive threshold segmentation algorithm（ATSA） based on image radius was developed and utilized to obtain the binary images of aggregates, air-voids and asphalt mastic. Then the shape and distribution of air-voids and aggregates were analyzed. The results show that the ATSA can distinguish the target and background effectively. Gradation and coarse aggregate size of asphalt mixtures have an obvious influence on the distribution of air-voids. The movements of aggregate particles are complex and aggregates with elliptic sharp show great rotation. The effect of gradation on microstructure during high-temperature damage promotes the research about the failure mechanism of asphalt concrete pavement.
