针对目前城市道路场景中行道树提取方法需要设置的参数较多以及树冠点云相互重叠难以精确分割的问题,文章采用一种行道树提取与单株木分割算法。首先通过布料滤波算法从原始点云中移除地面点,并利用半径滤波滤除离群点,去除地面点和噪...针对目前城市道路场景中行道树提取方法需要设置的参数较多以及树冠点云相互重叠难以精确分割的问题,文章采用一种行道树提取与单株木分割算法。首先通过布料滤波算法从原始点云中移除地面点,并利用半径滤波滤除离群点,去除地面点和噪声点对行道树提取的影响;然后通过增加PointNet++网络的点集抽象模块(set abstraction,SA)提高模型特征提取能力,使模型更适用于行道树点云的提取,并利用改进后的网络从原始点云中提取行道树点云;最后结合密度聚类算法(density-based spatial clustering of applications with noise,DBSCAN)与K-Means算法对相互重叠的行道树点云进行分割,得到单株木信息。为验证该方法的有效性,以北京永昌路道路数据集进行训练测试。结果表明:改进后模型的行道树点云平均提取精度和交并比(intersection over union,IoU)分别提高了9.2%和15.1%,达到了94.5%、0.916;单木分割平均精度达到了91.3%。展开更多
In the dispersed computing environment driven by intelligent networks,intrusion detection faces significant challenges.This paper proposes a multilayer decentralized federated learning framework based on mean field ga...In the dispersed computing environment driven by intelligent networks,intrusion detection faces significant challenges.This paper proposes a multilayer decentralized federated learning framework based on mean field game theory(MFG-DFL).The framework organizes networked computing points(NCPs)into a three-layer collaborative architecture,and innovatively introduces MFG theory to model the complex dynamic interactions,which among large-scale NCPs as a game between a representative NCP and the mean field.By solving the coupled HJB and FPK equations,we design a dynamic incentive mechanism to fairly quantify and reward NCP contributions,thus aligning individual rationality with the global objectives of the system.The simulation results on the CICIoT2023 data set demonstrate the outstanding performance of the proposed framework.Specifically,it achieves an intrusion detection accuracy of 81.09%in highly non-IID scenarios,showcasing a well-balanced trade-off between computational efficiency and performance enhancement.展开更多
Self-compassion is an important psychological resource that improves health-related quality of life and diabetes self-management;however,the psychological processes underlying these associations remain unclear.This st...Self-compassion is an important psychological resource that improves health-related quality of life and diabetes self-management;however,the psychological processes underlying these associations remain unclear.This study examined the roles of meaning in life and resilience in the relationship between self-compassion and both health-related quality of life and diabetes self-management.Participants were 301 individuals living with type 2 diabetes(176(58.5%)females,M age=49.69,SD=12.36)conveniently selected from two tertiary healthcare institutions in Nigeria.They completed self-report measures of self-compassion,meaning in life,resilience,health-related quality of life and diabetes selfmanagement.Multiple regression analyses indicated that self-compassion positively predicted health-related quality of life and diabetes self-management.Meaning in life and resilience independently and positively predicted health-related quality of life and diabetes self-management.Mediation analysis showed that meaning in life mediated the association between self-compassion and health-related quality of life,as well as diabetes self-management.Likewise,resilience mediated the connection between self-compassion and health-related quality of life,and diabetes self-management behaviour.Intervention for diabetes self-management should focus on promoting self-compassion,meaning in life and resilience abilities of patients,as they can potentially improve health and self-care behaviours needed for recovery.展开更多
文摘针对目前城市道路场景中行道树提取方法需要设置的参数较多以及树冠点云相互重叠难以精确分割的问题,文章采用一种行道树提取与单株木分割算法。首先通过布料滤波算法从原始点云中移除地面点,并利用半径滤波滤除离群点,去除地面点和噪声点对行道树提取的影响;然后通过增加PointNet++网络的点集抽象模块(set abstraction,SA)提高模型特征提取能力,使模型更适用于行道树点云的提取,并利用改进后的网络从原始点云中提取行道树点云;最后结合密度聚类算法(density-based spatial clustering of applications with noise,DBSCAN)与K-Means算法对相互重叠的行道树点云进行分割,得到单株木信息。为验证该方法的有效性,以北京永昌路道路数据集进行训练测试。结果表明:改进后模型的行道树点云平均提取精度和交并比(intersection over union,IoU)分别提高了9.2%和15.1%,达到了94.5%、0.916;单木分割平均精度达到了91.3%。
基金National Science Foundation Project of China(62436004,62372317)National Key Research and Development Program of China(2023YFC3331702)。
文摘In the dispersed computing environment driven by intelligent networks,intrusion detection faces significant challenges.This paper proposes a multilayer decentralized federated learning framework based on mean field game theory(MFG-DFL).The framework organizes networked computing points(NCPs)into a three-layer collaborative architecture,and innovatively introduces MFG theory to model the complex dynamic interactions,which among large-scale NCPs as a game between a representative NCP and the mean field.By solving the coupled HJB and FPK equations,we design a dynamic incentive mechanism to fairly quantify and reward NCP contributions,thus aligning individual rationality with the global objectives of the system.The simulation results on the CICIoT2023 data set demonstrate the outstanding performance of the proposed framework.Specifically,it achieves an intrusion detection accuracy of 81.09%in highly non-IID scenarios,showcasing a well-balanced trade-off between computational efficiency and performance enhancement.
文摘Self-compassion is an important psychological resource that improves health-related quality of life and diabetes self-management;however,the psychological processes underlying these associations remain unclear.This study examined the roles of meaning in life and resilience in the relationship between self-compassion and both health-related quality of life and diabetes self-management.Participants were 301 individuals living with type 2 diabetes(176(58.5%)females,M age=49.69,SD=12.36)conveniently selected from two tertiary healthcare institutions in Nigeria.They completed self-report measures of self-compassion,meaning in life,resilience,health-related quality of life and diabetes selfmanagement.Multiple regression analyses indicated that self-compassion positively predicted health-related quality of life and diabetes self-management.Meaning in life and resilience independently and positively predicted health-related quality of life and diabetes self-management.Mediation analysis showed that meaning in life mediated the association between self-compassion and health-related quality of life,as well as diabetes self-management.Likewise,resilience mediated the connection between self-compassion and health-related quality of life,and diabetes self-management behaviour.Intervention for diabetes self-management should focus on promoting self-compassion,meaning in life and resilience abilities of patients,as they can potentially improve health and self-care behaviours needed for recovery.
