Analysis and design of delta operator systems with nested actuator saturation

In this paper, the problem of estimating the domain of attraction is considered for delta operator systems subject to nested actuator saturation. A set invariance condition is established for the delta operator system with nested actuator saturation in terms of auxiliary feedback matrices. Based on the set invariance condition, an optimisation approach is proposed to estimate the domain of attraction for the delta operator system. Thereby, the partial results of nested actuator saturation for both continuous-time systems and discrete-time systems are extended to delta operator system framework. A numerical example is provided to illustrate the effectiveness of the proposed design techniques.     

Fault-tolerant control of delta operator systems with actuator saturation and effectiveness loss

This paper studies the problem of robust fault-tolerant control against the actuator effectiveness loss for delta operator systems with actuator saturation. Ellipsoids are used to estimate the domain of attraction for the delta operator systems with actuator saturation and effectiveness loss. Some invariance set conditions used for enlarging the domain of attraction are expressed by linear matrix inequalities. Discussions on system performance optimisation are presented in this paper, including reduction on computational complexity, expansion of the domain of attraction and disturbance rejection. Two numerical examples are given to illustrate the effectiveness of the developed techniques.     

一种融合Hough变换和ASM定位瞳孔中心点方法

特征点的准确定位是机器视觉和模式识别的关键技术之一.主动形状模型(ASM)是一种传统的图像特征点定位算法,具有较高的精确性和鲁棒性.为了提高ASM人脸瞳孔特征点定位的精确度,提出了使用Hough变换方法来改进瞳孔特征点的定位.通过ASM算法,初步定位出瞳孔特征点,并使用Sobel算子对图像进行边缘检测,然后在人眼位置选择一个合适的窗口使用Hough圆检测,找出精确瞳孔点相对于ASM初步定位瞳孔点的偏移量.在实验室采集的人脸图像上的对比实验表明,该方法能够显著的改善ASM人眼瞳孔特征点定位准确性.由于使用了初定位进行了搜索范围的限制,计算量也得到了有效的控制.     

基于JADE的自动聚类算法

JADE算法是传统差分进化算法（DE）的一种改进算法,其收敛速度更快、优化性能更好,拥有一套完整的自适应参数调整机制有效加强了算法的全局搜索优化能力. 本文将自适应差分进化算法（JADE）用于聚类,提出了一个新的基于JADE的自动聚类算法（AC-JADE）. 首先,本文采用双交叉策略,在传统的两点式交叉操作之后,针对DE用于自动聚类时的特定的编码方式,添加了一种基于个体间聚类中心随机交换交叉策略;其次,针对聚类中心选取方法的随机性导致的聚类中心有可能偏离数据集或者聚类中心过于集中的缺陷做出了相关改进,通过先对聚类中心进行筛选在进行聚类,有效避免了因算法本身的随机性导致的错误聚类划分. 通过对UCI的4个数据集的仿真实验比较,该种双交叉操作的聚类算法明显好于同类算法.     

An improved fruit fly optimization algorithm for solving the multidimensional knapsack problem

This paper presents an improved fruit fly optimization algorithm (IFFOA) for solving the multidimensional knapsack problem (MKP). In IFFOA, the parallel search is employed to balance exploitation and exploration. To make full use of swarm intelligence, a modified harmony search algorithm (MHS) is proposed and applied to add cooperation among swarms in IFFOA. In MHS, novel pitch adjustment scheme and random selection rule are developed by considering specific characters of MKP and FOA. Moreover, a vertical crossover is designed to guide stagnant dimensions out of local optima and further improve the performance. Extensive numerical simulations are conducted and comparisons with other state-of-the-art algorithms verify that the proposed algorithm is an effective alternative for solving the MKP.     

HACH: Heuristic Algorithm for Clustering Hierarchy protocol in wireless sensor networks

Wireless sensor networks (WSNs) require energy management protocols to efficiently use the energy supply constraints of battery-powered sensors to prolong its network lifetime. This paper proposes a novel Heuristic Algorithm for Clustering Hierarchy (HACH), which sequentially performs selection of inactive nodes and cluster head nodes at every round. Inactive node selection employs a stochastic sleep scheduling mechanism to determine the selection of nodes that can be put into sleep mode without adversely affecting network coverage. Also, the clustering algorithm uses a novel heuristic crossover operator to combine two different solutions to achieve an improved solution that enhances the distribution of cluster head nodes and coordinates energy consumption in WSNs. The proposed algorithm is evaluated via simulation experiments and compared with some existing algorithms. Our protocol shows improved performance in terms of extended lifetime and maintains favourable performances even under different energy heterogeneity settings.     

结合HSV与纹理特征的视频阴影消除算法

目的 在视频监控目标检测应用中,场景中的阴影会直接影响目标检测的准确度,因此阴影抑制算法研究显得尤为重要。目前广泛使用的是HSV（hue,saturation,value）阴影抑制方法,但是该方法存在由于亮度比值的阈值不稳定而造成将运动目标也检测为阴影的问题。针对该问题,本文提出了一种结合HSV与纹理特征的视频阴影消除方法。方法 首先将输入的图像使用传统的混合高斯模型建立背景并在灰度空间中提取前景,其次在HSV空间使用亮度比的阈值方法检测阴影,二者综合得到运动目标;针对由于亮度比值的阈值不稳定而导致的前景误检为阴影的问题,采用了LBP（local binary pattern）算子结合大津阈值（OTSU）提取部分运动目标。最后将LBP算子结合大津阈值提取的部分运动目标与HSV空间检测的目标两者相或,最终去除运动目标的阴影。结果 本文选用在CVPR-ATON和CAVIAR标准视频库中多个场景的阴影视频,将本文算法与SNP算法、SP算法、DNM1算法和DNM2算法进行对比仿真,实验结果表明本文算法在阴影检测率和阴影识别率的平均值上提升约10%。结论 本文提出的视频阴影消除算法结合了HSV与纹理特征,可以在不同的环境中有效地去除阴影,运动目标保留完整,可适用于智能视频监控、遥感图像和人机交互中。     

基于八方向Sobel算子的边缘检测算法

针对传统图像边缘检测方法检测效果不理想的问题,提出一种基于八方向Sobel算子的边缘检测算法。采用0°、22.5°、45°、67.5°、90°、112.5°、135°、157.5° 8个方向的模板进行检测,能较好地检测出不同方向的边缘。在检测过程中,考虑到邻域内像素到中心像素距离不同,对中心像素的贡献不同,算法 根据该像素到中心像素欧氏距离对邻域内像素进行加权,使得距中心像素越近,其权值越大。实验表明,算法检测出的图像边缘较为完整,轮廓线清晰且连续性较好。     

Time-local formulation and identification of implicit Volterra models by use of diffusive representation

We present a time-continuous identification method for nonlinear dynamic Volterra models of the form HX=f(u,X)+v with H, a causal convolution operator. It is mainly based on a suitable parameterization of H deduced from the so-called diffusive representation, which is devoted to state representations of integral operators. Following this approach, the complex dynamic nature of H can be summarized by a few numerical parameters on which the identification of the dynamic part of the model will focus. The method is validated on a physical numerical example.