Regular inspection of bridge cracks is crucial to bridge maintenance and repair.The traditional manual crack detection methods are timeconsuming,dangerous and subjective.At the same time,for the existing mainstream vi...Regular inspection of bridge cracks is crucial to bridge maintenance and repair.The traditional manual crack detection methods are timeconsuming,dangerous and subjective.At the same time,for the existing mainstream vision-based automatic crack detection algorithms,it is challenging to detect fine cracks and balance the detection accuracy and speed.Therefore,this paper proposes a new bridge crack segmentationmethod based on parallel attention mechanism and multi-scale features fusion on top of the DeeplabV3+network framework.First,the improved lightweight MobileNetv2 network and dilated separable convolution are integrated into the original DeeplabV3+network to improve the original backbone network Xception and atrous spatial pyramid pooling(ASPP)module,respectively,dramatically reducing the number of parameters in the network and accelerates the training and prediction speed of the model.Moreover,we introduce the parallel attention mechanism into the encoding and decoding stages.The attention to the crack regions can be enhanced from the aspects of both channel and spatial parts and significantly suppress the interference of various noises.Finally,we further improve the detection performance of the model for fine cracks by introducing a multi-scale features fusion module.Our research results are validated on the self-made dataset.The experiments show that our method is more accurate than other methods.Its intersection of union(IoU)and F1-score(F1)are increased to 77.96%and 87.57%,respectively.In addition,the number of parameters is only 4.10M,which is much smaller than the original network;also,the frames per second(FPS)is increased to 15 frames/s.The results prove that the proposed method fits well the requirements of rapid and accurate detection of bridge cracks and is superior to other methods.展开更多
The special channels and intrinsic defects within GO laminates make it a very potential candidate for gas separation in recent years. Herein, the gas separation performance of GO membranes prepared on the surface of c...The special channels and intrinsic defects within GO laminates make it a very potential candidate for gas separation in recent years. Herein, the gas separation performance of GO membranes prepared on the surface of ceramicα-Al_2O_3 hollow fibre was investigated systematically. The microstructures of ceramic hollow fibre supported GO membranes were optimized by adjusting operation conditions. And, the GO membrane fabricated at 30 min exhibited great promising H_2 recovery ability from H_2/CO_2 mixture. At room temperature, the H_2 permeance was over 1.00 × 10^(-7)mol·m^(-2)·s^(-1)·Pa^(-1)for both single gas and binary mixture. The corresponding ideal selectivity and mixture separation factor reached around 15 and 10, respectively. In addition, humility, operation temperature, H_2 concentration in the feed and the reproducibility were also studied in this work.展开更多
Objective：To evaluate the clinical value with positron emission tomography/computerized tomography（PET/CT） imaging for the detection of vulnerable plaque in atherosclerotic lesions. Methods：Sixty people with a age...Objective：To evaluate the clinical value with positron emission tomography/computerized tomography（PET/CT） imaging for the detection of vulnerable plaque in atherosclerotic lesions. Methods：Sixty people with a age of over 60[mean age （69.2 ± 7.1）years] underwent three dimension（3D） whole-body fluorine-18-2-fluoro-2-deoxy-D-glucose（^18F-FDG） PET/CT imaging and were evaluated retrospectively, including 6 cases assessed as normal and 54 cases with active atherosclerotic plaque. Fifty-four cases with SUVs and CT values in the aortic wall of high-FDG-uptake were measured retrospectively. These high-FDG-uptake cases in the aortic wall were divided into three groups according their CT value. Cases in group 1 had high uptake in atherosclerotic lesions of the aortic wall with CT value of less than 60 Hu（soft plaque）. Cases in group 2 had high uptake with CT value between 60-100 Hu （intermediate plaque）, Cases in group 3 had high uptake with CT value more than 100 Hu（calcified plaque）, Group 4 was normal. Results： In group 1, there were 42 high-FDG-uptake sites （average SUV 1.553 ± 0.486）. In group 2, there were 30 high-FDG-uptake sites（average SUV 1.393 ± 0.296）. In group 3, there were 36 high-FDG-uptake sites（average SUV 1.354 ± 0.189）. In group 4, there were 33 normal-FDG-uptake sites （average SUV was 1.102 ± 0.141）, The SUVs showed significant difference among the four groups（F = 678.909, P = 0.000）. There were also significant difference found between the normal-FDG-uptake group and the high-FDG-uptake groups（P = 0.000, 0.000, 0.001, respectively）. Conclusion：Different degrees of ^18F-FDG uptake in active large atherosclerotic plaque were shown in different stages of atherosclerotic plaque formation. The soft plaque had the highest FDG uptake in this study. This suggested that ^18F- FDG PET/CT imaging may be of great potential value in early diagnosis and monitoring of vulnerable soft plaque in atherosclerotic lesions.展开更多
基金This work was supported by the High-Tech Industry Science and Technology Innovation Leading Plan Project of Hunan Provincial under Grant 2020GK2026,author B.Y,http://kjt.hunan.gov.cn/.
文摘Regular inspection of bridge cracks is crucial to bridge maintenance and repair.The traditional manual crack detection methods are timeconsuming,dangerous and subjective.At the same time,for the existing mainstream vision-based automatic crack detection algorithms,it is challenging to detect fine cracks and balance the detection accuracy and speed.Therefore,this paper proposes a new bridge crack segmentationmethod based on parallel attention mechanism and multi-scale features fusion on top of the DeeplabV3+network framework.First,the improved lightweight MobileNetv2 network and dilated separable convolution are integrated into the original DeeplabV3+network to improve the original backbone network Xception and atrous spatial pyramid pooling(ASPP)module,respectively,dramatically reducing the number of parameters in the network and accelerates the training and prediction speed of the model.Moreover,we introduce the parallel attention mechanism into the encoding and decoding stages.The attention to the crack regions can be enhanced from the aspects of both channel and spatial parts and significantly suppress the interference of various noises.Finally,we further improve the detection performance of the model for fine cracks by introducing a multi-scale features fusion module.Our research results are validated on the self-made dataset.The experiments show that our method is more accurate than other methods.Its intersection of union(IoU)and F1-score(F1)are increased to 77.96%and 87.57%,respectively.In addition,the number of parameters is only 4.10M,which is much smaller than the original network;also,the frames per second(FPS)is increased to 15 frames/s.The results prove that the proposed method fits well the requirements of rapid and accurate detection of bridge cracks and is superior to other methods.
基金Supported by the National Natural Science Foundation of China(21476107,21490585,21406107)the Innovative Research Team Program by the Ministry of Education of China(IRT13070)the Topnotch Academic Programs Project of Jiangsu Higher Education Institutions(TAPP)
文摘The special channels and intrinsic defects within GO laminates make it a very potential candidate for gas separation in recent years. Herein, the gas separation performance of GO membranes prepared on the surface of ceramicα-Al_2O_3 hollow fibre was investigated systematically. The microstructures of ceramic hollow fibre supported GO membranes were optimized by adjusting operation conditions. And, the GO membrane fabricated at 30 min exhibited great promising H_2 recovery ability from H_2/CO_2 mixture. At room temperature, the H_2 permeance was over 1.00 × 10^(-7)mol·m^(-2)·s^(-1)·Pa^(-1)for both single gas and binary mixture. The corresponding ideal selectivity and mixture separation factor reached around 15 and 10, respectively. In addition, humility, operation temperature, H_2 concentration in the feed and the reproducibility were also studied in this work.
文摘Objective：To evaluate the clinical value with positron emission tomography/computerized tomography（PET/CT） imaging for the detection of vulnerable plaque in atherosclerotic lesions. Methods：Sixty people with a age of over 60[mean age （69.2 ± 7.1）years] underwent three dimension（3D） whole-body fluorine-18-2-fluoro-2-deoxy-D-glucose（^18F-FDG） PET/CT imaging and were evaluated retrospectively, including 6 cases assessed as normal and 54 cases with active atherosclerotic plaque. Fifty-four cases with SUVs and CT values in the aortic wall of high-FDG-uptake were measured retrospectively. These high-FDG-uptake cases in the aortic wall were divided into three groups according their CT value. Cases in group 1 had high uptake in atherosclerotic lesions of the aortic wall with CT value of less than 60 Hu（soft plaque）. Cases in group 2 had high uptake with CT value between 60-100 Hu （intermediate plaque）, Cases in group 3 had high uptake with CT value more than 100 Hu（calcified plaque）, Group 4 was normal. Results： In group 1, there were 42 high-FDG-uptake sites （average SUV 1.553 ± 0.486）. In group 2, there were 30 high-FDG-uptake sites（average SUV 1.393 ± 0.296）. In group 3, there were 36 high-FDG-uptake sites（average SUV 1.354 ± 0.189）. In group 4, there were 33 normal-FDG-uptake sites （average SUV was 1.102 ± 0.141）, The SUVs showed significant difference among the four groups（F = 678.909, P = 0.000）. There were also significant difference found between the normal-FDG-uptake group and the high-FDG-uptake groups（P = 0.000, 0.000, 0.001, respectively）. Conclusion：Different degrees of ^18F-FDG uptake in active large atherosclerotic plaque were shown in different stages of atherosclerotic plaque formation. The soft plaque had the highest FDG uptake in this study. This suggested that ^18F- FDG PET/CT imaging may be of great potential value in early diagnosis and monitoring of vulnerable soft plaque in atherosclerotic lesions.
