Objective: To investigate the epidemiological characteristics and relative factor about hospital death in patients with road traffic trauma. Methods: The age, sex, road-use category, sites of injury, injury severity s...Objective: To investigate the epidemiological characteristics and relative factor about hospital death in patients with road traffic trauma. Methods: The age, sex, road-use category, sites of injury, injury severity scale of 159 hospital death cases in 2436 cases with road traffic trauma were observed, and the relation between the causes of death and time elapsed after injury was also studied with likelihood ratio Chi-square test.展开更多
目的现有显著性检测方法大多只关注显著目标的中心信息,使得算法只能得到中心清晰、边缘模糊的显著目标,丢失了一些重要的边界信息,而使用核范数约束进行低秩矩阵恢复,运算过程冗余。为解决以上问题,本文提出一种无监督迭代重加权最小...目的现有显著性检测方法大多只关注显著目标的中心信息,使得算法只能得到中心清晰、边缘模糊的显著目标,丢失了一些重要的边界信息,而使用核范数约束进行低秩矩阵恢复,运算过程冗余。为解决以上问题,本文提出一种无监督迭代重加权最小二乘低秩恢复算法,用于图像视觉显著性检测。方法将图像分为细中粗3种尺度的分割,从细粒度和粗粒度先验的融合中得到分割先验信息;将融合后的分割先验信息通过迭代重加权最小二乘法求解平滑低秩矩阵恢复,生成粗略显著图;使用中粒度分割先验对粗略显著图进行平滑,生成最终的视觉显著图。结果实验在MSRA10K(Microsoft Research Asia 10K)、SOD(salient object detection dataset)和ECSSD(extended complex scene saliency dataset)数据集上进行测试,并与现有的11种算法进行对比。结果表明,本文算法可生成边界清晰的显著图。在MSRA10K数据集上,本文算法实现了最高的AUC(area under ROC(receiver operating characteristic)curve)和F-measure值,MAE(mean absolute error)值仅次于SMD(structured matrix decomposition)算法和RBD(robust back ground detection)算法,AUC和F-measure值比次优算法RPCA(robust principal component analysis)分别提高了3.9%和12.3%;在SOD数据集上,综合AUC、F-measure和MAE值来看,本文算法优于除SMD算法以外的其他算法,AUC值仅次于SMD算法、SC(smoothness constraint)算法和GBVS(graph-based visual salieney)算法,F-measure值低于最优算法SMD 2.6%;在ECSSD数据集上,本文算法实现了最高的F-measure值75.5%,AUC值略低于最优算法SC 1%,MAE值略低于最优算法HCNs(hierarchical co-salient object detection via color names)2%。结论实验结果表明,本文算法能从前景复杂或背景复杂的显著图像中更准确地检测出边界清晰的显著目标。展开更多
文摘Objective: To investigate the epidemiological characteristics and relative factor about hospital death in patients with road traffic trauma. Methods: The age, sex, road-use category, sites of injury, injury severity scale of 159 hospital death cases in 2436 cases with road traffic trauma were observed, and the relation between the causes of death and time elapsed after injury was also studied with likelihood ratio Chi-square test.
文摘目的现有显著性检测方法大多只关注显著目标的中心信息,使得算法只能得到中心清晰、边缘模糊的显著目标,丢失了一些重要的边界信息,而使用核范数约束进行低秩矩阵恢复,运算过程冗余。为解决以上问题,本文提出一种无监督迭代重加权最小二乘低秩恢复算法,用于图像视觉显著性检测。方法将图像分为细中粗3种尺度的分割,从细粒度和粗粒度先验的融合中得到分割先验信息;将融合后的分割先验信息通过迭代重加权最小二乘法求解平滑低秩矩阵恢复,生成粗略显著图;使用中粒度分割先验对粗略显著图进行平滑,生成最终的视觉显著图。结果实验在MSRA10K(Microsoft Research Asia 10K)、SOD(salient object detection dataset)和ECSSD(extended complex scene saliency dataset)数据集上进行测试,并与现有的11种算法进行对比。结果表明,本文算法可生成边界清晰的显著图。在MSRA10K数据集上,本文算法实现了最高的AUC(area under ROC(receiver operating characteristic)curve)和F-measure值,MAE(mean absolute error)值仅次于SMD(structured matrix decomposition)算法和RBD(robust back ground detection)算法,AUC和F-measure值比次优算法RPCA(robust principal component analysis)分别提高了3.9%和12.3%;在SOD数据集上,综合AUC、F-measure和MAE值来看,本文算法优于除SMD算法以外的其他算法,AUC值仅次于SMD算法、SC(smoothness constraint)算法和GBVS(graph-based visual salieney)算法,F-measure值低于最优算法SMD 2.6%;在ECSSD数据集上,本文算法实现了最高的F-measure值75.5%,AUC值略低于最优算法SC 1%,MAE值略低于最优算法HCNs(hierarchical co-salient object detection via color names)2%。结论实验结果表明,本文算法能从前景复杂或背景复杂的显著图像中更准确地检测出边界清晰的显著目标。