针对高速公路隧道内光线昏暗、图像受灯光影响及远距离小目标检测困难等问题,提出了一种改进的YOLOv5高速公路隧道车辆和人员检测算法。首先,使用高斯混合聚类来获得更加匹配数据集目标的一组锚框,提高了模型的检测精度;其次,在特征融...针对高速公路隧道内光线昏暗、图像受灯光影响及远距离小目标检测困难等问题,提出了一种改进的YOLOv5高速公路隧道车辆和人员检测算法。首先,使用高斯混合聚类来获得更加匹配数据集目标的一组锚框,提高了模型的检测精度;其次,在特征融合部分引入内容感知重组特征(content-aware ReAssembly of FEatures, CARAFE)上采样算子,扩大感受野,降低上采样过程特征细节损失;最后,通过向网络中插入坐标注意力(coordinate attention, CA),进一步增强模型对图像各位置特征的提取能力。为验证算法的有效性,在浙江温丽高速公路隧道数据集上进行实验,结果表明:所提算法的平均检测精度(mean average precision, mAP)达到了95.7%,较原模型提升3.8%,对于远距离小目标和受严重灯光影响的目标能够实现更加精准检测,为复杂环境下高速公路隧道内车辆和人员检测提供了一种有效的解决方案。展开更多
Under classical particle dynamics, the interaction process between intense femtosecond laser pulses and icosahedral noble-gas atomic clusters was studied. Our calculated results show that ionization proceeds mainly th...Under classical particle dynamics, the interaction process between intense femtosecond laser pulses and icosahedral noble-gas atomic clusters was studied. Our calculated results show that ionization proceeds mainly through tunnel ionization in the combined field from ions, electrons and laser, rather than the electron-impact ionization. With increasing cluster size, the average and maximum kinetic energy of the product ion increases. According to our calculation, the expansion process of the clusters after laser irradiation is dominated by Coulomb explosion and the expansion scale increases with increasing cluster size. The dependence of average kinetic energy and average charge state of the product ions on laser wavelength is also presented and discussed. The dependence of average kinetic energy on the number of atoms inside the cluster was studied and compared with the experimental data. Our results agree with the experimental results reasonably well.展开更多
We report the formation and local electronic structure of Ge clusters on the Si（111）-7×7 surface studied by using variable temperature scanning tunnelling microscopy （VT-STM） and low-temperature scanning tunn...We report the formation and local electronic structure of Ge clusters on the Si（111）-7×7 surface studied by using variable temperature scanning tunnelling microscopy （VT-STM） and low-temperature scanning tunnelling spectroscopy （STS）. Atom-resolved STM images reveal that the Ce atoms are prone to forming clusters with 1.0 nm in diameter for coverage up to 0.12 ML. Such Ce clusters preferentially nucleate at the centre of the faulted-half unit cells, leading to the ＇dark sites＇ of Si centre adatoms from the surrounding three unfaulted-half unit cells in filled-state images. Biasdependent STM images show the charge transfer from the neighbouring Si adatoms to Ce clusters. Low-temperature STS of the Ce clusters reveals that there is a band gap on the Ce cluster and the large voltage threshold is about 0.9 V.展开更多
Estimation of tunnel diameter convergence is a very important issue for tunneling construction,especially when the new Austrian tunneling method（NATM） is adopted.For this purpose,a systematic convergence measurement...Estimation of tunnel diameter convergence is a very important issue for tunneling construction,especially when the new Austrian tunneling method（NATM） is adopted.For this purpose,a systematic convergence measurement is usually implemented to adjust the design during the whole construction,and consequently deadly hazards can be prevented.In this study,a new fuzzy model capable of predicting the diameter convergences of a high-speed railway tunnel was developed on the basis of adaptive neuro-fuzzy inference system（ANFIS） approach.The proposed model used more than 1 000 datasets collected from two different tunnels,i.e.Daguan tunnel No.2 and Yaojia tunnel No.1,which are part of a tunnel located in Hunan Province,China.Six Takagi-Sugeno fuzzy inference systems were constructed by using subtractive clustering method.The data obtained from Daguan tunnel No.2 were used for model training,while the data from Yaojia tunnel No.1 were employed to evaluate the performance of the model.The input parameters include surrounding rock masses（SRM） rating index,ground engineering conditions（GEC） rating index,tunnel overburden（H）,rock density（？）,distance between monitoring station and working face（D）,and elapsed time（T）.The model’s performance was assessed by the variance account for（VAF）,root mean square error（RMSE）,mean absolute percentage error（MAPE） as well as the coefficient of determination（R2） between measured and predicted data as recommended by many researchers.The results showed excellent prediction accuracy and it was suggested that the proposed model can be used to estimate the tunnel convergence and convergence velocity.展开更多
文摘针对高速公路隧道内光线昏暗、图像受灯光影响及远距离小目标检测困难等问题,提出了一种改进的YOLOv5高速公路隧道车辆和人员检测算法。首先,使用高斯混合聚类来获得更加匹配数据集目标的一组锚框,提高了模型的检测精度;其次,在特征融合部分引入内容感知重组特征(content-aware ReAssembly of FEatures, CARAFE)上采样算子,扩大感受野,降低上采样过程特征细节损失;最后,通过向网络中插入坐标注意力(coordinate attention, CA),进一步增强模型对图像各位置特征的提取能力。为验证算法的有效性,在浙江温丽高速公路隧道数据集上进行实验,结果表明:所提算法的平均检测精度(mean average precision, mAP)达到了95.7%,较原模型提升3.8%,对于远距离小目标和受严重灯光影响的目标能够实现更加精准检测,为复杂环境下高速公路隧道内车辆和人员检测提供了一种有效的解决方案。
基金Projects supported by the National Natural Science Foundation of China (Grant Nos 10575046 and 10775062)
文摘Under classical particle dynamics, the interaction process between intense femtosecond laser pulses and icosahedral noble-gas atomic clusters was studied. Our calculated results show that ionization proceeds mainly through tunnel ionization in the combined field from ions, electrons and laser, rather than the electron-impact ionization. With increasing cluster size, the average and maximum kinetic energy of the product ion increases. According to our calculation, the expansion process of the clusters after laser irradiation is dominated by Coulomb explosion and the expansion scale increases with increasing cluster size. The dependence of average kinetic energy and average charge state of the product ions on laser wavelength is also presented and discussed. The dependence of average kinetic energy on the number of atoms inside the cluster was studied and compared with the experimental data. Our results agree with the experimental results reasonably well.
基金Project supported by the National Natural Science Foundation of China （Grant Nos 90406022 and 10674159）.
文摘We report the formation and local electronic structure of Ge clusters on the Si（111）-7×7 surface studied by using variable temperature scanning tunnelling microscopy （VT-STM） and low-temperature scanning tunnelling spectroscopy （STS）. Atom-resolved STM images reveal that the Ce atoms are prone to forming clusters with 1.0 nm in diameter for coverage up to 0.12 ML. Such Ce clusters preferentially nucleate at the centre of the faulted-half unit cells, leading to the ＇dark sites＇ of Si centre adatoms from the surrounding three unfaulted-half unit cells in filled-state images. Biasdependent STM images show the charge transfer from the neighbouring Si adatoms to Ce clusters. Low-temperature STS of the Ce clusters reveals that there is a band gap on the Ce cluster and the large voltage threshold is about 0.9 V.
基金support of China University of Geosciences (Wuhan)
文摘Estimation of tunnel diameter convergence is a very important issue for tunneling construction,especially when the new Austrian tunneling method（NATM） is adopted.For this purpose,a systematic convergence measurement is usually implemented to adjust the design during the whole construction,and consequently deadly hazards can be prevented.In this study,a new fuzzy model capable of predicting the diameter convergences of a high-speed railway tunnel was developed on the basis of adaptive neuro-fuzzy inference system（ANFIS） approach.The proposed model used more than 1 000 datasets collected from two different tunnels,i.e.Daguan tunnel No.2 and Yaojia tunnel No.1,which are part of a tunnel located in Hunan Province,China.Six Takagi-Sugeno fuzzy inference systems were constructed by using subtractive clustering method.The data obtained from Daguan tunnel No.2 were used for model training,while the data from Yaojia tunnel No.1 were employed to evaluate the performance of the model.The input parameters include surrounding rock masses（SRM） rating index,ground engineering conditions（GEC） rating index,tunnel overburden（H）,rock density（？）,distance between monitoring station and working face（D）,and elapsed time（T）.The model’s performance was assessed by the variance account for（VAF）,root mean square error（RMSE）,mean absolute percentage error（MAPE） as well as the coefficient of determination（R2） between measured and predicted data as recommended by many researchers.The results showed excellent prediction accuracy and it was suggested that the proposed model can be used to estimate the tunnel convergence and convergence velocity.
