The problem of the unmanned surface vessel (USV) path planning in static and dynamic obstacle environments is addressed in this paper. Multi-behavior fusion based potential field method is proposed, which contains thr...The problem of the unmanned surface vessel (USV) path planning in static and dynamic obstacle environments is addressed in this paper. Multi-behavior fusion based potential field method is proposed, which contains three behaviors: goal-seeking, boundary-memory following and dynamic-obstacle avoidance. Then, different activation conditions are designed to determine the current behavior. Meanwhile, information on the positions, velocities and the equation of motion for obstacles are detected and calculated by sensor data. Besides, memory information is introduced into the boundary following behavior to enhance cognition capability for the obstacles, and avoid local minima problem caused by the potential field method. Finally, the results of theoretical analysis and simulation show that the collision-free path can be generated for USV within different obstacle environments, and further validated the performance and effectiveness of the presented strategy.展开更多
为应对医疗数据面临的内外部爬取威胁,构建了一套基于用户和实体行为分析(User and Entity Behavior Analytics,UEBA)的协同防护体系。该体系通过建立用户与实体的动态行为基线,并融合流量异常、异地登录、高频访问等多维度特征,实现了...为应对医疗数据面临的内外部爬取威胁,构建了一套基于用户和实体行为分析(User and Entity Behavior Analytics,UEBA)的协同防护体系。该体系通过建立用户与实体的动态行为基线,并融合流量异常、异地登录、高频访问等多维度特征,实现了对外部恶意数据爬取与内部数据窃取行为的有效识别。在医院OA、互联网挂号及医院信息系统(Hospital Information System, HIS)中的实践证明,该体系成功识别了多起外部渗透与内部违规事件,显著增强了系统对数据泄露风险的主动防御能力。证明基于UEBA的防护体系可系统化地应对医疗场景下的数据爬取威胁,为智慧医院信息安全建设提供可复制、可推广的实践路径。展开更多
通过日常巡检对肉鸡异常状态进行及时识别,是提升集约化养殖管理效率的重要手段。相比传统人工巡检方式,基于计算机视觉的自动化巡检在检测效率和一致性方面具有明显优势,但在实际养殖环境中,肉鸡个体密集分布,小目标、多尺度变化及遮...通过日常巡检对肉鸡异常状态进行及时识别,是提升集约化养殖管理效率的重要手段。相比传统人工巡检方式,基于计算机视觉的自动化巡检在检测效率和一致性方面具有明显优势,但在实际养殖环境中,肉鸡个体密集分布,小目标、多尺度变化及遮挡现象普遍存在,给视觉检测模型的稳定应用带来挑战。针对上述问题,本研究基于YOLO11n(you only look once)模型,提出了一种改进的目标检测方法 GMA-YOLO11n(GSConv and multi-scale attention YOLO11n)。该模型在Backbone中引入GSConv轻量化卷积模块以降低计算复杂度;并通过多尺度特征融合新增160×160的高分辨率特征层,以增强对小尺度和密集目标的检测能力;同时在多尺度特征输入前引入SE(squeeze-and-excitation)通道注意力模块,提升关键特征表达。试验结果表明,该模型能够有效实现肉鸡饮水、进食、行走等日常行为及异常状态的多类别检测,在数据集Ⅰ和数据集Ⅱ上的平均精度均值mAP分别达到93.87%和90.45%,较基线模型均有所提升,且推理速度满足实际视频巡检需求。展开更多
基金financially supported by the National Natural Science Foundation of China(Grant No.51879049)DK-I Dynamic Positioning System Console Project
文摘The problem of the unmanned surface vessel (USV) path planning in static and dynamic obstacle environments is addressed in this paper. Multi-behavior fusion based potential field method is proposed, which contains three behaviors: goal-seeking, boundary-memory following and dynamic-obstacle avoidance. Then, different activation conditions are designed to determine the current behavior. Meanwhile, information on the positions, velocities and the equation of motion for obstacles are detected and calculated by sensor data. Besides, memory information is introduced into the boundary following behavior to enhance cognition capability for the obstacles, and avoid local minima problem caused by the potential field method. Finally, the results of theoretical analysis and simulation show that the collision-free path can be generated for USV within different obstacle environments, and further validated the performance and effectiveness of the presented strategy.
文摘为应对医疗数据面临的内外部爬取威胁,构建了一套基于用户和实体行为分析(User and Entity Behavior Analytics,UEBA)的协同防护体系。该体系通过建立用户与实体的动态行为基线,并融合流量异常、异地登录、高频访问等多维度特征,实现了对外部恶意数据爬取与内部数据窃取行为的有效识别。在医院OA、互联网挂号及医院信息系统(Hospital Information System, HIS)中的实践证明,该体系成功识别了多起外部渗透与内部违规事件,显著增强了系统对数据泄露风险的主动防御能力。证明基于UEBA的防护体系可系统化地应对医疗场景下的数据爬取威胁,为智慧医院信息安全建设提供可复制、可推广的实践路径。
文摘通过日常巡检对肉鸡异常状态进行及时识别,是提升集约化养殖管理效率的重要手段。相比传统人工巡检方式,基于计算机视觉的自动化巡检在检测效率和一致性方面具有明显优势,但在实际养殖环境中,肉鸡个体密集分布,小目标、多尺度变化及遮挡现象普遍存在,给视觉检测模型的稳定应用带来挑战。针对上述问题,本研究基于YOLO11n(you only look once)模型,提出了一种改进的目标检测方法 GMA-YOLO11n(GSConv and multi-scale attention YOLO11n)。该模型在Backbone中引入GSConv轻量化卷积模块以降低计算复杂度;并通过多尺度特征融合新增160×160的高分辨率特征层,以增强对小尺度和密集目标的检测能力;同时在多尺度特征输入前引入SE(squeeze-and-excitation)通道注意力模块,提升关键特征表达。试验结果表明,该模型能够有效实现肉鸡饮水、进食、行走等日常行为及异常状态的多类别检测,在数据集Ⅰ和数据集Ⅱ上的平均精度均值mAP分别达到93.87%和90.45%,较基线模型均有所提升,且推理速度满足实际视频巡检需求。
