为提高目标形变、遮挡、相似干扰以及视野超出等复杂场景的目标跟踪性能,提出一种复杂场景单目标跟踪算法。基于Staple算法研究二维高斯函数像素权重赋予问题,优化颜色直方图统计,增强目标与背景区分度。引入基于峰值旁瓣比(Peak Side L...为提高目标形变、遮挡、相似干扰以及视野超出等复杂场景的目标跟踪性能,提出一种复杂场景单目标跟踪算法。基于Staple算法研究二维高斯函数像素权重赋予问题,优化颜色直方图统计,增强目标与背景区分度。引入基于峰值旁瓣比(Peak Side Lobe Ratio,PSR)的HOG特征、颜色特征的自适应融合机制,合理选择融合系数,确保混合特征更加可靠。分析目标区域中心与上一帧目标中心距离,结合最大混合响应计算最佳中心位置,解决相似目标干扰问题。采用混合响应、HOG特征、平均峰值相关能量(Average Peak-to-Correlation Energy,APCE)判定目标丢失、遮挡情况,保持目标框位置,实现目标的及时重新跟踪。采用结合之前帧和当前帧信息的模板更新策略,进一步提升跟踪精度,并在OTB100数据集中涉及形变、遮挡、视野超出3个属性视频上测试。实验结果表明,改进算法在整体和特定属性(形变、遮挡、出视野)的成功率及形变属性的精确度上,较Staple算法分别提升了1.8%,3.3%,2%和9%;在VOT16数据集上,改进算法在整体和遮挡属性上,重叠度较Staple提升了0.0222和0.0196,满足复杂的特定场景下的目标跟踪需求。展开更多
The problem of the stability analysis and controller design which the network-induced delays and data dropout problems network-induced delays are assumed to be time-varying and bounded, for Lurie networked control sys...The problem of the stability analysis and controller design which the network-induced delays and data dropout problems network-induced delays are assumed to be time-varying and bounded, for Lurie networked control systems （NCSs） is investigated, in are simultaneously considered. By considering that the and analyzing the relationship between the delay and its upper bound, employing a Lyapunov-Krasovskii function and an integral inequality approach, an improved stability criterion for NCSs is proposed. Furthermore, the resulting condition is extended to design a less conservative state feedback controller by employing an improved cone complementary linearization （ICCL） algorithm. Numerical examples are provided to show the effectiveness of the method.展开更多
基金Project(61025015)supported by the National Natural Science Foundation of China for Distinguished Young ScholarsProject (IRT1044)supported by the Program for Changjiang Scholars and Innovative Research Team in University of China+2 种基金Projects(61143004,61203136,61074067,61273185)supported by the National Natural Science Foundation of ChinaProjects(12JJ4062,11JJ2033)supported by the Natural Science Foundation of Hunan Province,ChinaProject(12C0078)supported by Hunan Provincial Department of Education,China
文摘The problem of the stability analysis and controller design which the network-induced delays and data dropout problems network-induced delays are assumed to be time-varying and bounded, for Lurie networked control systems （NCSs） is investigated, in are simultaneously considered. By considering that the and analyzing the relationship between the delay and its upper bound, employing a Lyapunov-Krasovskii function and an integral inequality approach, an improved stability criterion for NCSs is proposed. Furthermore, the resulting condition is extended to design a less conservative state feedback controller by employing an improved cone complementary linearization （ICCL） algorithm. Numerical examples are provided to show the effectiveness of the method.
