第四章讨论了改进的卡尔曼跟踪滤波器。
The fourth chapter is about kalman trackers and an improvement.
采用卡尔曼滤波器对目标进行跟踪时，目标初始状态估计是影响初始阶段跟踪精度的一个重要原因。
When using Kalman Filter to track a target, estimation of the initial state of the target is an important factor influencing tracking precision in the initial phase.
针对基于卡尔曼滤波器的残差检验法会跟踪软故障的局限性，提出了一种基于状态递推器的改进方法。
For the disadvantage of tracking the soft failure of the residual test based on Kalman filter, this paper makes an improvement on the method by using the state propagator.
针对线性随机系统提出了一种改进强跟踪卡尔曼滤波器(MSTKF)。
A modified strong tracking Kalman filter (MSTKF) for linear stochastic systems is proposed.
在毫米波雷达目标跟踪中，角闪烁的非高斯特性将使得经典的卡尔曼滤波器失效。
In MMW radar tracking, the classical Kalman filter will degrade seriously when observation noise is non-Gaussian because of target glint.
由于选择了新的机动加速度量，从而得出线性的状态方程，由机动指令的实时估计得到机动目标自适应跟踪卡尔曼滤波器。
A linear state equation is got from selection of maneuvering acceleration. The adaptivity of adaptive tracking Kalman filter is represented by estimation of maneuvering commander at real time.
跟踪阶段用卡尔曼滤波器结合肤色特征跟踪人脸，如果跟踪失败，转入检测阶段。
In the tracking stage, track face using kalman filter and skin-color feature, if fail to track then turn into detecting stage.
为了有效解决运动目标遮挡时目标信息容易丢失从而导致跟踪失败的问题，提出一种基于卡尔曼滤波器的运动目标跟踪算法。
In order to effectively solve the problem that the loss of object information under occlusion causes the failure of tracking, moving objects tracking algorithm is presented based on Kalman filter.
在以常规卡尔曼滤波器为基础的各种跟踪算法中，要求精确的模型和噪声统计，但在实际问题中，大多数情况上述要求不能满足。
Accurate models and noise statistics are required in many tracking algorithms based on the traditional Kalman filter, which are difficult to be satisfied in engineering application.
为了提高系统图像处理速度，利用卡尔曼滤波器对跟踪的特征点进行预测，并用窗口处理技术减小图像处理区域。
In order to improve the speed of image processing, the Kalman filter is used to predict the next place of tracked point.
模板的元素取自目标特征值的概率，通过48个卡尔曼滤波器可以跟踪所有特征值的概率变化。
The element of template is probability of eigenvalue of target. These probability are acquire by a kalman filter group which had 48 kalman filters.
结果表明：恒定增益卡尔曼滤波器在跟踪精度，对目标机动的响应能力以及可实现性等指标上都具有较好的特性。
The results show that the tracking accuracy, the responsibility to target mobility and the implementation of steady Kalman filter are good.
在水下被动目标跟踪系统中，直角坐标系下的扩展卡尔曼滤波器容易发散而导致滤波精度很差。
Concerning the problem to instability and low accuracy of the passive filter on bearings-only target tracking, a modified adaptive Extended Kalman filter algorithm on polar coordinate is presented.
仿真结果表明，对于纯方位跟踪问题，UPF不仅解决了扩展卡尔曼滤波器的线性化损失难题，而且与PF等粒子滤波器相比，具有更高的跟踪精度。
The results show that the UPF not only solves the linearized loss problem in the extended Kalman filter, but also is more accurate than the PF in the BOT.
仿真结果表明，该改进滤波器跟踪机动目标的精度高于常规卡尔曼滤波器和强跟踪卡尔曼滤波器。
The results of simulation indicate that this new approach has a better accuracy than the traditional Kalman filter and strong tracking Kalman filter in the field of maneuvering target tracking.
该无源被动定位系统采用了加权最小二乘法来估计目标位置并用卡尔曼滤波器对目标进行跟踪与预测。
Weighted least square estimation for target location and Kalman filtering for tracking and predication are used in such passive location system.
本文根据随机逼近原理提出了一种自适应卡尔曼滤波器，适用于被动声纳对水下噪声源的跟踪。
A self-adapting Kalman Filter based on the principle of stochastic approximatation is introduced in the paper. This filter provides an ideal means for the tracking of underwater noise source.
针对卡尔曼滤波器对系统模型依赖性强、鲁棒性差和跟踪机动目标能力有限的问题，提出了一种新的利用混合模糊逻辑和标准卡尔曼滤波器的联合算法。
The Kalman filter has been commonly used in target tracking, however its performance may be degraded in presence of maneuver, low robustness and strong model dependence.
然后将目标跟踪技术应用于特征点的跟踪，利用两个一维卡尔曼滤波器分别对特征点的两个坐标进行预测跟踪。
Then the object tracking technique is introduced into the tracking of feature points, using two 1d Kalman filter to track the two coordinates of feature points.
然后将目标跟踪技术应用于特征点的跟踪，利用两个一维卡尔曼滤波器分别对特征点的两个坐标进行预测跟踪。
Then the object tracking technique is introduced into the tracking of feature points, using two 1d Kalman filter to track the two coordinates of feature points.
应用推荐