动力定位冗余测量系统的模糊自适应融合算法

针对船舶动力定位的冗余测量系统,基于模糊自适应滤波和滤波器可信度的模糊评判,建立了多传感器数据融合算法。该方法与线性或非线性系统状态估计相结合,构成模糊自适应融合算法。通过监测滤波新息的均值和分歧度,应用模糊推理系统在线调节测量噪声协方差矩阵,从而抑制滤波发散;构建新的数据质量函数和子系统故障检测函数,根据检测结果自适应地选择子系统构建全局融合体系;利用子系统滤波的新息分歧度比值和状态估计误差的协方差,模糊评判各子系统滤波的可信度,由此计算相应的信息分配系数,实现全局融合。通过转台半实物仿真,验证了该算法的有效性。     

点云模型基于几何特征增强与保特征曲面拟合的法向估计方法EI北大核心CSCD

现有法向估计中,尤其是模型中存在较大噪声的情况下,目标点邻域的选择是一个关键且困难的问题.针对点云模型,为了提高法向估计准确度,提出一种自适应选择邻域且保特征、抗噪声的法向估计方法.首先,提出双边非局部特征增强模块,根据网络前置学习特征以及邻域点几何特性对点邻域进行加权选择,并据此对局部特征进行增强,以提升网络对模型局部几何特征学习的能力;然后,采用局部特征与全局特征相结合的形式刻画点云完备的几何特征,并以此为基础进行局部曲面拟合及法向估计;最后,在局部曲面拟合中提出邻域保特征损失,依据邻域点受噪声干扰度对邻域点的拟合权重进行调整,实现保细节特征的局部曲面拟合,提高对噪声的鲁棒性.实验使用PCPNET数据集进行模型训练和测试,大量定性与定量的实验结果表明,与相关方法相比,所提方法对于不同噪声级别以及不同密度分布等复杂情形都可取得更加准确的法向估计结果,并更好地推动曲面重建等点云处理应用.     

基于视差空间的双目视觉里程计

提出了一种基于视差空间的双目视觉里程计算法.利用SIFT特征点的尺度和旋转不变性,实现左、右图像对特征点的准确匹配,及前后帧间的特征跟踪.在RANSAC框架下对匹配点进行运动估计获得运动参数初始值,然后迭代更新匹配点的视差比值直至收敛.为克服传统算法中3维空间噪声分布不均匀的缺陷,利用了视差空间噪声分布的各向同性的性质进行运动估计,并且通过迭代取得全局最小值.实验结果表明,该算法在运动估计中具有更好的精度.     

一种基于局部描述符的三维人脸识别方法

提出一种局部描述符进行三维人脸识别。每个采样点的局部特征定义为该点根据其法向量与3个主轴之间的角度自适应选取的邻域点集向人脸主轴平面投影所得的面积。文中提出的三维人脸识别算法首先对人脸进行预处理,归一化到较统一的姿态后,提取与鼻尖等距的轮廓线,并对轮廓线进行重采样以剔除无用点。然后对每个采样点提取局部特征。最后建立人脸之间的点对应关系,将加权融合后的局部特征用于识别。通过实验认证,文中方法识别效果较好,且对遮挡和噪声有较好的鲁棒性。     

融合局部特征的面部遮挡表情识别

目的 针对人脸表情识别中存在局部遮挡的问题,提出一种融合局部特征的面部遮挡表情识别方法。方法 首先,为了减少噪声的影响,利用高斯滤波对归一化后的图像进行去噪处理;然后根据人脸不同部位对表情识别的不同贡献度,将图像划分为两个重要的子区域,并分别对该子区域进行不重叠分块处理;采用改进的中心对称局部二值模式（差值中心对称局部二值模式DCS-LBP）和改进的差值局部方向模式（梯度中心对称局部方向模式GCS-LDP）对各个子块提取相应的特征,并采用级联的方式得到图像的特征直方图;最后结合最近邻分类器对表情图像进行分类识别：利用卡方距离求取测试集图像与训练集图像特征直方图之间的距离,同时考虑到遮挡的干扰以及每个子块包含信息量的不同,利用信息熵对子块得到的卡方距离进行自适应加权。结果 在日本女性人脸表情库（JAFFE）和Cohn-Kanade（CK）人脸表情库上进行了3次交叉实验。在JAFFE库中随机遮挡、嘴部遮挡和眼部遮挡分别可以取得92.86%、94.76%和86.19%以上的平均识别率;在CK库中随机遮挡、嘴部遮挡和眼部遮挡分别可以取得99%、98.67%和99%以上的平均识别率。结论 该特征提取方法通过融合梯度方向上灰度值的差异以及梯度方向之间边缘响应值的差异来描述图像的特征,更加完整地提取了图像的细节信息。针对遮挡情况,本文采用的图像分割和信息熵自适应加权方法,有效地降低了遮挡对表情识别的干扰。在相同的实验环境下,与经典的局部特征提取方法以及遮挡问题处理方法的对比表明了该方法的有效性和优越性。     

基于投票决策的实时遮挡处理技术

针对当前基于深度信息的虚实遮挡处理技术面临的实时性差和精度低的问题,提出一种基于局部区域深度估计和基于patch相似性噪声点投票融合的实时虚实遮挡处理算法.该算法将真实场景视频序列作为输入,首先利用局部区域深度估计算法通过稀疏重建估算出稀疏关键点的深度信息,对稀疏深度施加目标区域的约束限制深度向周围像素的传播,从而快速...     

应用于状态监测的多传感器融合估计

在状态监测的工程实际中, 使用多个同类传感器进行在线测量可以得到更为准确的状态估计．但各传感器测量噪声会出现相关的情况, 而且很难得到相关测量噪声的方差矩阵的精确值, 测量系统往往是不确定的．本文根据系统测量将系统分解为确定和不确定扰动两部分, 分别进行估计, 然后将两者的融合估计结果相加得到了最优鲁棒的融合估计．针对确定部分, 利用同类传感器的测量方差为Pei-Radman矩阵的特性, 通过求解测量噪声方差矩阵的最大特征值得到了一种简便的最优融合估计算法, 该算法避免了求解方差矩阵的逆的过程．针对不确定     

复杂场景下基于C-SHOT特征的3D物体识别与位姿估计

为了准确地同时识别复杂点云中的多个目标,提出一种基于法矢改进点云特征C-SHOT的3D物体识别方法.首先,在估计RGB-D数据的点云法矢时将邻域点距离信息考虑在内,计算带距离权重的协方差矩阵得到更精确的点云法矢;其次根据特征点处法矢与邻域法矢的夹角余弦构造点云形状直方图,同时统计点云纹理直方图并与形状直方图融合成C-SHOT描述符;最后对场景与模板分别提取C-SHOT特征,利用Kd树快速求得对应对,引入3D霍夫投票机制,并结合点云局部坐标系克服噪声遮挡问题完成多目标初识别.基于LM-ICP实现精确定位及位姿估计,画出目标包围盒,采用基准数据库CVLab以及采集实验室真实数据进行实验,结果验证了该方法的有效性与精确性.     

多速率传感器不完全测量数据自适应融合算法

针对不完全测量的多速率传感器系统,建立了自适应的分级融合算法。该算法构建了传感器子系统故障检测方法,检测子系统故障,确认不完全测量,避免单点故障污染整个系统;建立子系统的切换控制逻辑,引入映射矩阵,构成新的信息分享系数,自适应地选择子系统参与中心融合;对于不完全测量信息的传感器子系统,利用子滤波预测值代替估计值进行融合;全局状态估计利用各子系统的当前与前一时刻的估计值,提高了融合性能。对建立的算法进行仿真,验证了该算法的有效性。     

基于自适应噪声抵消的MIMO-OFDM信道估计

对多输入多输出正交频分复用系统的信道估计进行了研究,利用自适应噪声抵消的思想,提出了一种基于自适应噪声抵消和奇异值分解的信道估计方法.该方法首先利用自适应噪声抵消对最小二乘(LS)估计进行去噪处理,然后将其结果用于奇异值分解(SVD)的MMSE低阶近似方法中.仿真结果表明,利用此方法有效地抑制了噪声的影响,其系统性能优于利用最小二乘方法所获得的系统性能.     

通信受限下网络化多传感器系统的Kalman融合估计

研究了一类通信受限下网络化多传感器系统的 Kalman 融合估计问题, 其中通信受限 是指系统在一个采样周期内只允许有限个传感器与融合中心通信. 首先, 提出了一种周期性分组传输的通信策略, 并将每组传感器所对应的局部估计系统描述成一个离散周期子系统模型. 其次, 每个子系统根据最新测量信息的更新时刻, 选择相应的 Kalman 估计器 (滤波器或预报器), 从而得到各子系统在每一时刻的一个局部最优估计, 再通过矩阵加权线性最小方差最优融合准则得到最优融合估计,并给出了Kalman融合估计器的设计方法. 最后, 通过一个目标跟踪例子验证所提方法的有效性.     

抗遮挡的自适应运动目标跟踪方法

针对复杂场景中的目标遮挡问题,提出一种基于均值漂移(Mean shift)和轨迹校正的自适应目标跟踪方法.由于Mean shift迭代易陷入局部最优点,这里引入Kalman滤波器以预测和校正目标运动轨迹,并根据迭代轨迹误差校正协方差,使得跟踪器在多峰值非高斯分布的复杂环境下也能收敛到全局最优点.基于Bhattacharrya系数计算色彩x、y方向分量相似度,并根据邻帧分量相似度偏差自适应调整相似度融合权值.综合当前帧和前面帧作用更新目标运动状态、特征和尺度模型.实验结果表明提出的方法对于静态场景遮挡和目标间互遮挡、部分和全部遮挡下的目标跟踪均具有鲁棒的跟踪性能.     

带未知通信干扰和丢包补偿的多传感器网络化不确定系统的分布式融合滤波

研究了带有未知通信干扰、观测丢失和乘性噪声不确定性的多传感器网络化系统的状态估计问题.通过白色乘性噪声描述系统状态和观测中的随机不确定性,采用一组服从Bernoulli分布的随机变量描述网络传输过程中存在的观测丢失现象,且数据传输中存在未知的网络通信干扰.当发生丢包时,以当前丢失观测的预报值进行补偿.对每个单传感器子系统,应用线性无偏最小方差估计准则设计了不依赖于未知通信干扰的最优线性滤波器.推导了任两个局部滤波误差之间的互协方差阵.进而,应用矩阵加权融合估计算法给出了分布式融合状态滤波器.仿真例子验证了算法的有效性.     

Robust fusion steady‐state filtering for multisensor networked systems with one‐step random delay,missing measurements,and uncertain‐variance multiplicative and additive white noises

The robust fusion steady‐state filtering problem is investigated for a class of multisensor networked systems with mixed uncertainties including multiplicative noises, one‐step random delay, missing measurements, and uncertain noise variances, the phenomena of one‐step random delay and missing measurements occur in a random way, and are described by two Bernoulli distributed random variables with known conditional probabilities. Using a model transformation approach, which consists of augmented approach, derandomization approach, and fictitious noise approach, the original multisensor system under study is converted into a multimodel multisensor system with only uncertain noise variances. According to the minimax robust estimation principle, based on the worst‐case subsystems with conservative upper bounds of uncertain noise variances, the robust local steady‐state Kalman estimators (predictor, filter, and smoother) are presented in a unified framework. Applying the optimal fusion algorithm weighted by matrices, the robust distributed weighted state fusion steady‐state Kalman estimators are derived for the considered system. In addition, by using the proposed model transformation approach, the centralized fusion system is obtained, furthermore the robust centralized fusion steady‐state Kalman estimators are proposed. The robustness of the proposed estimators is proved by using a combination method consisting of augmented noise approach, decomposition approach of nonnegative definite matrix, matrix representation approach of quadratic form, and Lyapunov equation approach, such that for all admissible uncertainties, the actual steady‐state estimation error variances of the estimators are guaranteed to have the corresponding minimal upper bounds. The accuracy relations among the robust local and fused steady‐state Kalman estimators are proved. An example with application to autoregressive signal processing is proposed, which shows that the robust local and fusion signal estimation problems can be solved by the state estimation problems. Simulation example verifies the effectiveness and correctness of the proposed results.     

Distributed weighted robust Kalman filter fusion for uncertain systems with autocorrelated and cross-correlated noises

In this paper, the problem of distributed weighted robust Kalman filter fusion is studied for a class of uncertain systems with autocorrelated and cross-correlated noises. The system under consideration is subject to stochastic uncertainties or multiplicative noises. The process noise is assumed to be one-step autocorrelated. For each subsystem, the measurement noise is one-step autocorrelated, and the process noise and the measurement noise are two-step cross-correlated. An optimal robust Kalman-type recursive filter is first designed for each subsystem. Then, based on the newly obtained optimal robust Kalman-type recursive filter, a distributed weighted robust Kalman filter fusion algorithm is derived for uncertain systems with multiple sensors. The distributed fusion algorithm involves a recursive computation of the filtering error cross-covariance matrix between any two subsystems. Compared with the centralized Kalman filter, the distributed weighted robust Kalman filter developed in this paper has stronger fault-tolerance ability. Simulation results are provided to demonstrate the effectiveness of the proposed approaches.     

UUV水下回收中的视觉和短基线定位融合

为了使UUV在水下坞舱回收过程中利用视觉和短基线(short baseline-SBL)进行导引定位,提出了一种视觉和短基线的自适应融合定位方法,以提高导引定位的精度.介绍了短基线定位和视觉定位2种定位系统及其工作原理,以及定位数据的野值剔除和去噪方法.野值剔除采用了一种基于数据变化率的自适应在线野值剔除方法,数据去噪采用了软阈值小波滤波方法.针对传统卡尔曼滤波进行数据融合时先验知识不足的缺点,提出了一种基于模糊逻辑的在线自适应卡尔曼滤波融合方法.通过获取的实时测量数据,实时调整噪声的协方差矩阵来融合2种定位数据.水下回收水池试验结果表明,定位传感器的绝大部分野值被剔除且去噪效果明显,视觉和短基线融合后的定位精度有很大提高,证明了所提方法的有效性.     

多传感器最优信息融合白噪声反卷积滤波器

应用现代时间序列分析方法和白噪声估计理论,基于线性最小方差意义下按标量加权最优信息融合准则,对于带白色和有色观测噪声的多传感器单通道系统,提出了分布式融合白噪声反卷积滤波器.它由局部白噪声反卷积滤波器加权构成.可统一处理融合滤波、平滑和预报问题.给出了计算局部滤波误差互协方差公式,可用于计算最优加权.同单传感器情形相比,可提高融合滤波器精度.它可应用于石油地震勘探信号处理.一个3传感器信息融合Bernou lli-Gaussian白噪声反卷积滤波器的仿真例子说明了其有效性.     

Human steering angle estimation in video based on key point detection and Kalman filter

Human pose recognition and estimation in video is pervasive. However, the process noise and local occlusion bring great challenge to pose recognition. In this paper, we introduce the Kalman filter into pose recognition to reduce noise and solve local occlusion problem. The core of pose recognition in video is the fast detection of key points and the calculation of human steering angles. Thus, we first build a human key point detection model. Frame skipping is performed based on the Hamming distance of the hash value of every two adjacent frames in video. Noise reduction is performed on key point coordinates with the Kalman filter. To calculate the human steering angle, current state information of key points is predicted using the optimal estimation of key points at the previous time. Then human steering angle can be calculated based on current and previous state information. The improved SENet, NLNet and GCNet modules are integrated into key point detection model for improving accuracy. Tests are also given to illustrate the effectiveness of the proposed algorithm.     

Robust weighted state fusion Kalman estimators for networked systems with mixed uncertainties

This paper is concerned with robust weighted state fusion estimation problem for a class of time-varying multisensor networked systems with mixed uncertainties including uncertain-variance multiplicative and linearly correlated additive white noises, and packet dropouts. By augmented state method and fictitious noise technique, the original system is converted into one with only uncertain noise variances. According to the minimax robust estimation principle, based on the worst-case system with the conservative upper bounds of uncertain noise variances, four weighted state fusion robust Kalman estimators (filter, predictor and smoother) are presented in a unified form that the robust filter and smoother are designed based on the robust Kalman predictor. Their robustness is proved by the Lyapunov equation approach in the sense that their actual estimation error variances are guaranteed to have the corresponding minimal upper bounds for all admissible uncertainties. Their accuracy relations are proved. The corresponding robust local and fused steady-state Kalman estimators are also presented, and the convergence in a realization between the time-varying and steady-state robust Kalman estimators is proved by the dynamic error system analysis (DESA) method. Finally, a simulation example applied to uninterruptible power system (UPS) shows the correctness and effectiveness of the proposed results.