Based on the second order random wave solutions of water wave equations in finite water depth, statistical distributions of the depth integrated local horizontal momentum components are derived by use of the charact...Based on the second order random wave solutions of water wave equations in finite water depth, statistical distributions of the depth integrated local horizontal momentum components are derived by use of the characteristic function expansion method. The parameters involved in the distributions can be all determined by the water depth and the wave number spectrum of ocean waves. As an illustrative example, a fully developed wind generated sea is considered and the parameters are calculated for typical wind speeds and water depths by means of the Donelan and Pierson spectrum. The effects of nonlinearity and water depth on the distributions are also investigated.展开更多
目的从单幅影像中估计景深已成为计算机视觉研究热点之一,现有方法常通过提高网络的复杂度回归深度,增加了数据的训练成本及时间复杂度,为此提出一种面向单目深度估计的多层次感知条件随机场模型。方法采用自适应混合金字塔特征融合策略...目的从单幅影像中估计景深已成为计算机视觉研究热点之一,现有方法常通过提高网络的复杂度回归深度,增加了数据的训练成本及时间复杂度,为此提出一种面向单目深度估计的多层次感知条件随机场模型。方法采用自适应混合金字塔特征融合策略,捕获图像中不同位置间的短距离和长距离依赖关系,从而有效聚合全局和局部上下文信息,实现信息的高效传递。引入条件随机场解码机制,以此精细捕捉像素间的空间依赖关系。结合动态缩放注意力机制增强对不同图像区域间依赖关系的感知能力,引入偏置学习单元模块避免网络陷入极端值问题,保证模型的稳定性。针对不同特征模态间的交互情况,通过层次感知适配器扩展特征映射维度增强空间和通道交互性能,提高模型的特征学习能力。结果在NYU Depth v2(New York University depth dataset version 2)数据集上进行消融实验,结果表明,本文网络可以显著提高性能指标,相较于对比的先进方法,绝对相对误差(absolute relative error,Abs Rel)减小至0.1以内,降低7.4%,均方根误差(root mean square error,RMSE)降低5.4%。为验证模型在真实道路环境中的实用性,在KITTI(Karlsruhe Institute of Technology and Toyota Technological Institute at Chicago)数据集上进行对比实验,上述指标均优于对比的主流方法,其中RMSE降低3.1%,阈值(δ<1.25^(2),δ<1.25^(3))准确度接近100%,此外,在MatterPort3D数据集上验证了模型的泛化性。从可视化实验结果看,在复杂环境下本文方法可以更好地估计困难区域的深度。结论本文采用多层次特征提取器及混合金字塔特征融合策略,优化了信息在编码器和解码器间的传递过程,通过全连接解码获取像素级别的输出,能够有效提高单目深度估计精度。展开更多
Concrete carbonation relates to the diffusion of CO2 inthe atmosphere through the concrete pores and to the dissolution of the hydrates. A great deal of uncertainty still remains in this subject as a result of the ran...Concrete carbonation relates to the diffusion of CO2 inthe atmosphere through the concrete pores and to the dissolution of the hydrates. A great deal of uncertainty still remains in this subject as a result of the random nature of the infiltration of these aggressive agents as well as the material and structural properties of reinforced concrete. Consequently, in order to determine the degradation of these structures with time, it is necessary to approach the subject from a probabilistic point of view that it considers the uncertainties involved with the geometrical dimensions of these structures, with the material characteristics and with the environment and the subsequent deterioration that it provokes. The aim of this paper is the evaluation of carbonation depth (Xc) from a probabilistic analysis, focusing specifically on the study of the randomness of the most influential parameters on the reinforced concrete carbonation.展开更多
BACKGROUND： Certain neural functions, such as peripheral reflexes, differ between genders, while higher brain functions, such as language, are asymmetrically distributed between the two hemispheres. The question rema...BACKGROUND： Certain neural functions, such as peripheral reflexes, differ between genders, while higher brain functions, such as language, are asymmetrically distributed between the two hemispheres. The question remains as to whether depth perception differs between hemispheric laterality and genders, and whether it is affected by eye dominance. OBJECTIVE： To determine whether depth perception is influenced by factors such as gender, eye dominance, and hemispheric lateralization by recording evoked potential associated with depth perception. DESIGN, TIME AND SETTING： A contrast observation based on neuroelectrophysiology was performed at the Department of Biophysics of Ege University Medical School between June 2006 and April 2007. PARTICIPANTS： A total of 34 subjects, 19 females and 15 males, were included in the study with a mean age of （31.0 ± 6.9） years. All subjects were free of neurological or psychological disorders, or problems such as strabismus or vision correction. METHODS： Random-dot stereograms were used to elicit brain activity. A specially designed signal acquisition system employing two computers was used to record evoked potentials from both hemispheres via two pairs of scalp electrodes placed over the occipitotemporal areas of both hemispheres at symmetrical locations. MAIN OUTCOME MEASURES： Negative potential with a mean latency of （211.21 ±25.55） ms and a mean amplitude of （6.05 ± 1.53） pV was recorded from both occipitotemporal areas in 30 out of 34 participants. This was termed ＂Nd＂ and represented the evoked potential associated with depth perception. RESULTS： There were no significant differences in Nd amplitude or latency between the two hemispheres, the two eyes, or genders （P 〉 0.05）. CONCLUSION： The evoked potential associated with depth perception was not influenced by gender, hemisphere, or eye dominance.展开更多
This paper deals the randomness effect of the pressure of carbonic gas on the carbonation phenomenon of the reinforced concrete. This analysis concentrates on the evaluation of carbonation depth (Xc) and the carbonati...This paper deals the randomness effect of the pressure of carbonic gas on the carbonation phenomenon of the reinforced concrete. This analysis concentrates on the evaluation of carbonation depth (Xc) and the carbonation time (T1) which is the time necessary so that the face of carbonation arrives until the reinforcement from a probabilistic analysis. Monte Carlo simulations are realized under the assumption that the carbonic gas on the surface of the concrete is random variable with a log-normal probability distribution.展开更多
近年来深度学习在短文本聚类方面发挥巨大作用,最近提出的短文本聚类(Short Text Clustering, STC)算法在此方面取得不错的成效。为进一步提高聚类准确率并优化算法性能,基于指数函数提出改进的随机近邻嵌入算法。该算法用指数函数度量...近年来深度学习在短文本聚类方面发挥巨大作用,最近提出的短文本聚类(Short Text Clustering, STC)算法在此方面取得不错的成效。为进一步提高聚类准确率并优化算法性能,基于指数函数提出改进的随机近邻嵌入算法。该算法用指数函数度量样本点与聚类中心差距,放大不同特征差别,并在后期使用k-means++算法预先确定聚类中心与聚类数目。在Stackoverflow数据集上的实验证明,随机指数嵌入聚类模型(e-STC)在准确率与标准互信息上均优于原STC模型,准确率相对提高3.2%,互信息相对提高2.9%。展开更多
文摘Based on the second order random wave solutions of water wave equations in finite water depth, statistical distributions of the depth integrated local horizontal momentum components are derived by use of the characteristic function expansion method. The parameters involved in the distributions can be all determined by the water depth and the wave number spectrum of ocean waves. As an illustrative example, a fully developed wind generated sea is considered and the parameters are calculated for typical wind speeds and water depths by means of the Donelan and Pierson spectrum. The effects of nonlinearity and water depth on the distributions are also investigated.
文摘目的从单幅影像中估计景深已成为计算机视觉研究热点之一,现有方法常通过提高网络的复杂度回归深度,增加了数据的训练成本及时间复杂度,为此提出一种面向单目深度估计的多层次感知条件随机场模型。方法采用自适应混合金字塔特征融合策略,捕获图像中不同位置间的短距离和长距离依赖关系,从而有效聚合全局和局部上下文信息,实现信息的高效传递。引入条件随机场解码机制,以此精细捕捉像素间的空间依赖关系。结合动态缩放注意力机制增强对不同图像区域间依赖关系的感知能力,引入偏置学习单元模块避免网络陷入极端值问题,保证模型的稳定性。针对不同特征模态间的交互情况,通过层次感知适配器扩展特征映射维度增强空间和通道交互性能,提高模型的特征学习能力。结果在NYU Depth v2(New York University depth dataset version 2)数据集上进行消融实验,结果表明,本文网络可以显著提高性能指标,相较于对比的先进方法,绝对相对误差(absolute relative error,Abs Rel)减小至0.1以内,降低7.4%,均方根误差(root mean square error,RMSE)降低5.4%。为验证模型在真实道路环境中的实用性,在KITTI(Karlsruhe Institute of Technology and Toyota Technological Institute at Chicago)数据集上进行对比实验,上述指标均优于对比的主流方法,其中RMSE降低3.1%,阈值(δ<1.25^(2),δ<1.25^(3))准确度接近100%,此外,在MatterPort3D数据集上验证了模型的泛化性。从可视化实验结果看,在复杂环境下本文方法可以更好地估计困难区域的深度。结论本文采用多层次特征提取器及混合金字塔特征融合策略,优化了信息在编码器和解码器间的传递过程,通过全连接解码获取像素级别的输出,能够有效提高单目深度估计精度。
文摘Concrete carbonation relates to the diffusion of CO2 inthe atmosphere through the concrete pores and to the dissolution of the hydrates. A great deal of uncertainty still remains in this subject as a result of the random nature of the infiltration of these aggressive agents as well as the material and structural properties of reinforced concrete. Consequently, in order to determine the degradation of these structures with time, it is necessary to approach the subject from a probabilistic point of view that it considers the uncertainties involved with the geometrical dimensions of these structures, with the material characteristics and with the environment and the subsequent deterioration that it provokes. The aim of this paper is the evaluation of carbonation depth (Xc) from a probabilistic analysis, focusing specifically on the study of the randomness of the most influential parameters on the reinforced concrete carbonation.
文摘BACKGROUND： Certain neural functions, such as peripheral reflexes, differ between genders, while higher brain functions, such as language, are asymmetrically distributed between the two hemispheres. The question remains as to whether depth perception differs between hemispheric laterality and genders, and whether it is affected by eye dominance. OBJECTIVE： To determine whether depth perception is influenced by factors such as gender, eye dominance, and hemispheric lateralization by recording evoked potential associated with depth perception. DESIGN, TIME AND SETTING： A contrast observation based on neuroelectrophysiology was performed at the Department of Biophysics of Ege University Medical School between June 2006 and April 2007. PARTICIPANTS： A total of 34 subjects, 19 females and 15 males, were included in the study with a mean age of （31.0 ± 6.9） years. All subjects were free of neurological or psychological disorders, or problems such as strabismus or vision correction. METHODS： Random-dot stereograms were used to elicit brain activity. A specially designed signal acquisition system employing two computers was used to record evoked potentials from both hemispheres via two pairs of scalp electrodes placed over the occipitotemporal areas of both hemispheres at symmetrical locations. MAIN OUTCOME MEASURES： Negative potential with a mean latency of （211.21 ±25.55） ms and a mean amplitude of （6.05 ± 1.53） pV was recorded from both occipitotemporal areas in 30 out of 34 participants. This was termed ＂Nd＂ and represented the evoked potential associated with depth perception. RESULTS： There were no significant differences in Nd amplitude or latency between the two hemispheres, the two eyes, or genders （P 〉 0.05）. CONCLUSION： The evoked potential associated with depth perception was not influenced by gender, hemisphere, or eye dominance.
文摘This paper deals the randomness effect of the pressure of carbonic gas on the carbonation phenomenon of the reinforced concrete. This analysis concentrates on the evaluation of carbonation depth (Xc) and the carbonation time (T1) which is the time necessary so that the face of carbonation arrives until the reinforcement from a probabilistic analysis. Monte Carlo simulations are realized under the assumption that the carbonic gas on the surface of the concrete is random variable with a log-normal probability distribution.
文摘近年来深度学习在短文本聚类方面发挥巨大作用,最近提出的短文本聚类(Short Text Clustering, STC)算法在此方面取得不错的成效。为进一步提高聚类准确率并优化算法性能,基于指数函数提出改进的随机近邻嵌入算法。该算法用指数函数度量样本点与聚类中心差距,放大不同特征差别,并在后期使用k-means++算法预先确定聚类中心与聚类数目。在Stackoverflow数据集上的实验证明,随机指数嵌入聚类模型(e-STC)在准确率与标准互信息上均优于原STC模型,准确率相对提高3.2%,互信息相对提高2.9%。
