http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Huamei Zhu,Zhihang Li,Mengqi Huang,Pengxuan Ji,Hongyu Huang,Qianbing Zhang 국제구조공학회 2023 Smart Structures and Systems, An International Jou Vol.31 No.4
Post-earthquake building condition assessment is crucial for subsequent rescue and remediation and can be automated by emerging computer vision and deep learning technologies. This study is based on an endeavour for the 2nd International Competition of Structural Health Monitoring (IC-SHM 2021). The task package includes five image segmentation objectives - defects (crack/spall/rebar exposure), structural component, and damage state. The structural component and damage state tasks are identified as the priority that can form actionable decisions. A multi-task Convolutional Neural Network (CNN) is proposed to conduct the two major tasks simultaneously. The rest 3 sub-tasks (spall/crack/rebar exposure) were incorporated as auxiliary tasks. By synchronously learning defect information (spall/crack/rebar exposure), the multi-task CNN model outperforms the counterpart single-task models in recognizing structural components and estimating damage states. Particularly, the pixel-level damage state estimation witnesses a mIoU (mean intersection over union) improvement from 0.5855 to 0.6374. For the defect detection tasks, rebar exposure is omitted due to the extremely biased sample distribution. The segmentations of crack and spall are automated by single-task U-Net but with extra efforts to resample the provided data. The segmentation of small objects (spall and crack) benefits from the resampling method, with a substantial IoU increment of nearly 10%.
One-step deep learning-based method for pixel-level detection of fine cracks in steel girder images
Huamei Zhu,Zhihang Li,Mengqi Huang,Pengxuan Ji,Qianbing Zhang 국제구조공학회 2022 Smart Structures and Systems, An International Jou Vol.29 No.1
Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.
UWB Radar Imaging of Multiple Targets through Multi-Layer Walls
Huamei Zhang,Yerong Zhang,Fangfang Wang 보안공학연구지원센터 2016 International Journal of Hybrid Information Techno Vol.9 No.8
In the practical application of Ultra Wideband (UWB) through-wall radar imaging, the walls that are generally encountered consist of multiple layers, and there may be multiple targets, which presents challenges for the imaging algorithm. In this study, both the back-projection (BP) algorithm, which is a time-domain algorithm, and the phase-shift migration (PSM) algorithm, which is a frequency-domain algorithm, are used to locate two targets behind a two-layer wall. The image quality, the azimuth and range resolutions, the image entropy and the processing time of the two algorithms are compared. The simulation results demonstrate that the BP algorithm produces better image quality and that the PSM algorithm is faster. The effect of the sampling interval on the image entropy and processing time is discussed. In addition, the feasibility and the validity of the two algorithms are tested under noisy conditions. The results show that the two algorithms are robust, regardless of the presence of noise.
( Huamei Zhang ),( Dongdong Li ),( Jinlong Zhao ),( Haitao Wang ) 한국정보처리학회 2017 Journal of information processing systems Vol.13 No.4
In order to solve the undetected probability of multiple targets in ultra-wideband (UWB) through-the-wall radar imaging (TWRI), a time-delay and amplitude modified back projection (BP) algorithm is proposed. The refraction point is found by Fermat`s principle in the presence of a wall, and the time-delay is correctly compensated. On this basis, transmission loss of the electromagnetic wave, the absorption loss of the refraction wave, and the diffusion loss of the spherical wave are analyzed in detail. Amplitude compensation is deduced and tested on a model with a single-layer wall. The simulating results by finite difference time domain (FDTD) show that it is effective in increasing the scattering intensity of the targets behind the wall. Compensation for the diffusion loss in the spherical wave also plays a main role. Additionally, the two-layer wall model is simulated. Then, the calculating time and the imaging quality are compared between a singlelayer wall model and a two-layer wall model. The results illustrate the performance of the time-delay and amplitude-modified BP algorithm with multiple targets and multiple-layer walls of UWB TWRI.
Huamei Duan,Yunxia Yang3,Jim Patel,Nick Burke,Yuchun Zhai,Paul A. Webley,Dengfu Chen,Mujun Long 한국탄소학회 2018 Carbon Letters Vol.25 No.-
Activated carbon (AC) was modified by ammonium persulphate or nitric acid, respectively. AC and the modified materials were used as catalyst supports. The oxygen groups were introduced in the supports during the modifications. All the supports were characterized by N2-physisorption, Raman, X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), and thermogravimetric analysis. Methanol synthesis catalysts were prepared through wet impregnation of copper nitrate and zinc nitrate on the supports followed by thermal decomposition. These catalysts were measured by the means of N2-physisorption, X-ray diffraction, XPS, temperature programmed reduction and TEM tests. The catalytic performances of the prepared catalysts were compared with a commercial catalyst (CZA) in this work. The results showed that the methanol production rate of AC-CZ (23 mmol-CH3OH/(g-Cu·h)) was higher, on Cu loading basis, than that of CZA (9 mmol-CH3OH/(g-Cu·h)). We also found that the modification methods produced strong metal-support interactions leading to poor catalytic performance. AC without any modification can prompt the catalytic performance of the resulted catalyst.
Zhang, Huamei,Li, Dongdong,Zhao, Jinlong,Wang, Haitao Korea Information Processing Society 2017 Journal of information processing systems Vol.13 No.4
In order to solve the undetected probability of multiple targets in ultra-wideband (UWB) through-the-wall radar imaging (TWRI), a time-delay and amplitude modified back projection (BP) algorithm is proposed. The refraction point is found by Fermat's principle in the presence of a wall, and the time-delay is correctly compensated. On this basis, transmission loss of the electromagnetic wave, the absorption loss of the refraction wave, and the diffusion loss of the spherical wave are analyzed in detail. Amplitude compensation is deduced and tested on a model with a single-layer wall. The simulating results by finite difference time domain (FDTD) show that it is effective in increasing the scattering intensity of the targets behind the wall. Compensation for the diffusion loss in the spherical wave also plays a main role. Additionally, the two-layer wall model is simulated. Then, the calculating time and the imaging quality are compared between a single-layer wall model and a two-layer wall model. The results illustrate the performance of the time-delay and amplitude-modified BP algorithm with multiple targets and multiple-layer walls of UWB TWRI.
Yongqiang Zhang,Xiangyi Xue,Jingli Zhang,Shewei Xin,Hao Pan,Huamei Sun,Huiming Li 대한금속·재료학회 2021 METALS AND MATERIALS International Vol.27 No.7
The hot deformation behavior of Ti–22Al–25Nb was studied by the high temperature compression over a range of temperatures (950–1050 °C) and strain rates (0.001–10 s−1) in this paper. The work-hardening (WH) and softening deformationbehaviors of Ti–22Al–25Nb were analyzed. Obvious linear decreasing regimes of WH rate curves can be found before thedynamic recrystallization (DRX) onset, which indicates WH+ DRV (dynamic recovery) stage. And WH rate decreasedsignifcantly with strain rate reduced and temperature elevated. A physically-based constitutive model was established,which can well predict the fow behavior of Ti–22Al–25Nb. Additional, strain-rate sensitivity coefcient distribution mapwas established. The higher values of m appeared at the low strain rate. When the strain rate exceeded 0.1 s−1, the values ofm were lower than 0.25.