The large-scale periodic orbits of a nonlinear mechanics system can represent the homology classes, which are generally non-trivial, of the energy level surface and the topology properties of an energy level surface a...The large-scale periodic orbits of a nonlinear mechanics system can represent the homology classes, which are generally non-trivial, of the energy level surface and the topology properties of an energy level surface are determined by the that of the phase space and the large-scale properties of the Hamiltonian. These properties are used for estimate of the rank of the first homology group of energy level surfaces in the paper.展开更多
针对由于血管类间具有强相似性造成的动静脉错误分类问题,提出了一种新的融合上下文信息的多尺度视网膜动静脉分类网络(multi-scale retinal artery and vein classification network,MCFNet),该网络使用多尺度特征(multi-scale feature...针对由于血管类间具有强相似性造成的动静脉错误分类问题,提出了一种新的融合上下文信息的多尺度视网膜动静脉分类网络(multi-scale retinal artery and vein classification network,MCFNet),该网络使用多尺度特征(multi-scale feature,MSF)提取模块及高效的全局上下文信息融合(efficient global contextual information aggregation,EGCA)模块结合U型分割网络进行动静脉分类,抑制了倾向于背景的特征并增强了血管的边缘、交点和末端特征,解决了段内动静脉错误分类问题。此外,在U型网络的解码器部分加入3层深度监督,使浅层信息得到充分训练,避免梯度消失,优化训练过程。在2个公开的眼底图像数据集(DRIVE-AV,LES-AV)上,与3种现有网络进行方法对比,该模型的F1评分分别提高了2.86、1.92、0.81个百分点,灵敏度分别提高了4.27、2.43、1.21个百分点,结果表明所提出的模型能够很好地解决动静脉分类错误的问题。展开更多
In object oriented paradigm, cohesion of a class refers to the degree to which members of the class are interrelated. Metrics have been defined to measure cohesiveness of a class both at design and source code levels....In object oriented paradigm, cohesion of a class refers to the degree to which members of the class are interrelated. Metrics have been defined to measure cohesiveness of a class both at design and source code levels. In comparison to source code level class cohesion metrics, only a few design level class cohesion metrics have been proposed. Design level class cohesion metrics are based on the assumption that if all the methods of a class have access to similar para-meter types then they all process closely related information. A class with a large number of parameter types common in its methods is more cohesive than a class with less number of parameter types common in its methods. In this paper, we review the design level class cohesion metrics with a special focus on metrics which use similarity of parameter types of methods of a class as the basis of its cohesiveness. Basically three metrics fall in this category: Cohesion among Methods of a Class (CAMC), Normalized Hamming Distance (NHD), and Scaled NHD (SNHD). Keeping in mind the anomalies in the definitions of the existing metrics, a variant of the existing metrics is introduced. It is named NHD Modified (NHDM). An automated metric collection tool is used to collect the metric data from an open source software program. The metric data is then subjected to statistical analysis.展开更多
文摘The large-scale periodic orbits of a nonlinear mechanics system can represent the homology classes, which are generally non-trivial, of the energy level surface and the topology properties of an energy level surface are determined by the that of the phase space and the large-scale properties of the Hamiltonian. These properties are used for estimate of the rank of the first homology group of energy level surfaces in the paper.
文摘针对由于血管类间具有强相似性造成的动静脉错误分类问题,提出了一种新的融合上下文信息的多尺度视网膜动静脉分类网络(multi-scale retinal artery and vein classification network,MCFNet),该网络使用多尺度特征(multi-scale feature,MSF)提取模块及高效的全局上下文信息融合(efficient global contextual information aggregation,EGCA)模块结合U型分割网络进行动静脉分类,抑制了倾向于背景的特征并增强了血管的边缘、交点和末端特征,解决了段内动静脉错误分类问题。此外,在U型网络的解码器部分加入3层深度监督,使浅层信息得到充分训练,避免梯度消失,优化训练过程。在2个公开的眼底图像数据集(DRIVE-AV,LES-AV)上,与3种现有网络进行方法对比,该模型的F1评分分别提高了2.86、1.92、0.81个百分点,灵敏度分别提高了4.27、2.43、1.21个百分点,结果表明所提出的模型能够很好地解决动静脉分类错误的问题。
文摘In object oriented paradigm, cohesion of a class refers to the degree to which members of the class are interrelated. Metrics have been defined to measure cohesiveness of a class both at design and source code levels. In comparison to source code level class cohesion metrics, only a few design level class cohesion metrics have been proposed. Design level class cohesion metrics are based on the assumption that if all the methods of a class have access to similar para-meter types then they all process closely related information. A class with a large number of parameter types common in its methods is more cohesive than a class with less number of parameter types common in its methods. In this paper, we review the design level class cohesion metrics with a special focus on metrics which use similarity of parameter types of methods of a class as the basis of its cohesiveness. Basically three metrics fall in this category: Cohesion among Methods of a Class (CAMC), Normalized Hamming Distance (NHD), and Scaled NHD (SNHD). Keeping in mind the anomalies in the definitions of the existing metrics, a variant of the existing metrics is introduced. It is named NHD Modified (NHDM). An automated metric collection tool is used to collect the metric data from an open source software program. The metric data is then subjected to statistical analysis.
