Object-based change detection has been the hotspot in remote sensing image processing.A new approach toward object-based change detection is proposed.The two different temporal images are unitedly segmented using the ...Object-based change detection has been the hotspot in remote sensing image processing.A new approach toward object-based change detection is proposed.The two different temporal images are unitedly segmented using the mean shift procedure to obtain corresponding objects.Then change detection is implemented based on the integration of corresponding objects’ intensity and texture differences.Experiments are conducted on both panchromatic images and multispectral images and the results show that the integrated measure is robust with respect to illumination changes and noise.Supplementary color detection is conducted to determine whether the color of the unchanged objects changes or not when dealing with multispectral images.Some verification work is carried out to show the accuracy of the proposed approach.展开更多
低速率拒绝服务(LDoS,low-rate denial of service)攻击是一种降质服务(RoQ,reduction of quality)攻击,具有平均速率低和隐蔽性强的特点,它是云计算平台和大数据中心面临的最大安全威胁之一。提取了LDoS攻击流量的3个内在特征,建立基...低速率拒绝服务(LDoS,low-rate denial of service)攻击是一种降质服务(RoQ,reduction of quality)攻击,具有平均速率低和隐蔽性强的特点,它是云计算平台和大数据中心面临的最大安全威胁之一。提取了LDoS攻击流量的3个内在特征,建立基于BP神经网络的LDoS攻击分类器,提出了基于联合特征的LDoS攻击检测方法。该方法将LDoS攻击的3个内在特征组成联合特征作为BP神经网络的输入,通过预先设定的决策指标,达到检测LDoS攻击的目的。采用LDoS攻击流量专用产生工具,在NS2仿真平台和test-bed网络环境中对检测算法进行了测试与验证,实验结果表明通过假设检验得出检测率为96.68%。与现有研究成果比较说明基于联合特征的LDoS攻击检测性优于单个特征,并具有较高的计算效率。展开更多
地震预警(Earthquake Early Warning,EEW)是一种有效的防震减灾手段,能够在破坏性地震动到达前向公众发布预警,从而减少人员伤亡和财产损失。近年来,美国ShakeAlert地震预警系统发展迅速,2017年已正式在西海岸地区投入使用。文中详细介...地震预警(Earthquake Early Warning,EEW)是一种有效的防震减灾手段,能够在破坏性地震动到达前向公众发布预警,从而减少人员伤亡和财产损失。近年来,美国ShakeAlert地震预警系统发展迅速,2017年已正式在西海岸地区投入使用。文中详细介绍了ShakeAlert系统的发展历史、系统架构、警报发布及示范应用情况,旨在讨论ShakeAlert在地震中发挥的防震减灾的巨大效用。展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.40601084,40801178)National High Technology Research & Development Program of China("863"Program)(Grant Nos.2007AA120203,2009AA12Z145)
文摘Object-based change detection has been the hotspot in remote sensing image processing.A new approach toward object-based change detection is proposed.The two different temporal images are unitedly segmented using the mean shift procedure to obtain corresponding objects.Then change detection is implemented based on the integration of corresponding objects’ intensity and texture differences.Experiments are conducted on both panchromatic images and multispectral images and the results show that the integrated measure is robust with respect to illumination changes and noise.Supplementary color detection is conducted to determine whether the color of the unchanged objects changes or not when dealing with multispectral images.Some verification work is carried out to show the accuracy of the proposed approach.
文摘低速率拒绝服务(LDoS,low-rate denial of service)攻击是一种降质服务(RoQ,reduction of quality)攻击,具有平均速率低和隐蔽性强的特点,它是云计算平台和大数据中心面临的最大安全威胁之一。提取了LDoS攻击流量的3个内在特征,建立基于BP神经网络的LDoS攻击分类器,提出了基于联合特征的LDoS攻击检测方法。该方法将LDoS攻击的3个内在特征组成联合特征作为BP神经网络的输入,通过预先设定的决策指标,达到检测LDoS攻击的目的。采用LDoS攻击流量专用产生工具,在NS2仿真平台和test-bed网络环境中对检测算法进行了测试与验证,实验结果表明通过假设检验得出检测率为96.68%。与现有研究成果比较说明基于联合特征的LDoS攻击检测性优于单个特征,并具有较高的计算效率。
文摘地震预警(Earthquake Early Warning,EEW)是一种有效的防震减灾手段,能够在破坏性地震动到达前向公众发布预警,从而减少人员伤亡和财产损失。近年来,美国ShakeAlert地震预警系统发展迅速,2017年已正式在西海岸地区投入使用。文中详细介绍了ShakeAlert系统的发展历史、系统架构、警报发布及示范应用情况,旨在讨论ShakeAlert在地震中发挥的防震减灾的巨大效用。
