在此基础上，使用动态图像处理技术中的二维运动估计技术，结合背景更新模型提出了一种车辆跟踪算法。
On this basis, combined with one sort of background updating model, vehicle tracking algorithm is presented through using two-dimension motion estimation technique in dynamic image processing.
分析了三种不同的凹函数预测模型，提出了一种适合硬件实现的运动估计快速半像素级搜索算法。
A fast half-pixel motion estimation algorithm is proposed after analyzing three different mathematical prediction models based on concave functions.
基于随机分形理论，采用分数布朗运动模型，应用R/S分析技术和基于小波变换的谱参数估计方法计算地震道信号的赫斯特指数。
The thesis based on random fractal theory, utilizing fractional Brownian motion model, using R/S analysis technique and wavelet transform method to estimate Hurst exponent which dep.
针对静态背景下的背景差法，通过研究如何得到、以及如何及时更新背景模型，增强运动目标检测随环境变化的鲁棒性，提出了多级分块的背景估计方法。
Background estimation based on block is proposed to solve the problem of building background model and updating background which can enhance the robust of moving object detection.
以人脸肤色模型为基础，结合目标形状特征识别方法，并用扩展卡尔曼滤波估计目标运动轨迹，实现基于肤色的人脸实时跟踪鲁棒方法。
The face motion could be estimated by using face skin color model integrated with feature based object recognition technique and extended Kalman filter.
借助于图像运动的变阶参数模型和鲁棒回归分析，提出一种基于运动分割的图像运动估计方法。
A motion based segmentation scheme for image motion estimation is proposed using variable order parameterized models of image motion and robust regression.
本文研究了背景相减算法、差分法、基于光流场的运动估计和基于主动形状模型的目标跟踪算法。
In the thesis, the algorithm of background subtracting, frame difference algorithm, motion estimation based on optical flow field and object tracking based on active shape contour are investigated.
这两种方法的共同点是都是通过建立目标运动的参数模型，获得目标的运动状态估计，从而实现目标跟踪。
Both tracking methods achieve the target tracking state in common by establishing the motion parameter model and obtaining the motion state estimation of the target.
提出了一种基于CAD模型的目标运动估计和目标轨迹实时生成方法。
A scheme of kinetic parameters estimation and real-time track occurrence of objects based on the CAD models is investigated.
针对实际应用的不同侧重点，本文核心内容包括三方面：LV的运动矢量场估计、LV参数化运动跟踪模型和LV的概率跟踪模型。
To the different problems in practice, three main aspects are involved in this paper: the estimation of motion vector field, the parametric tracking model of lv and probability tracking model of lv.
文章提出了一种水声网络探测节点的运动目标参数估计方法，即基于模型匹配原理的估计算法。
A moving target parameters estimation algorithm, which is based on the principle of model matching and used for underwater acoustic networks detection nodes, is proposed in this paper.
与MPEG 4验证模型中的形状运动估计算法以及现有的改进算法相比，该算法大大提高了形状运动估计的速度，而形状的编码效率并没有降低。
Compared with the ME algorithm for shape coding in MPEG-4 verification model and other improved algorithms, the proposed algorithm significantly speeds the ME process and takes equal bit rate.
此算法基于多分辨率运动估计方案和新率失真模型进行预测模式快速选择，并结合带提前退出准则的快速搜索算法来加速搜索过程。
The multi-resolution motion estimation scheme and an efficient rate-distortion model are adopted with early termination rules to accelerate the search process.
本文对H．263中基于平移模型的重叠块运动估计补偿方法进行改进，提出了一种基于仿射模型的重叠块运动补偿方法，该方法对复杂运动图像质量有明显改善。
This paper proposes an overlapped block affine motion compensation method to improve the translational motion model used in H. 263, and it improved the quality of image with complicated motion.
算法从全局运动估计的基础出发，利用背景宏块运动相似性的特点快速建立背景宏块集合并采用常用的四参数全局运动估计模型估计运动参数。
From this point, the algorithm is implemented by searching and constructing a background image efficiently. With regard to global motion models, four-parameter estimation model is adopted.
在充分研究现有运动目标检测算法的基础上，提出了一种新的非参数核密度估计背景模型。
A new background model of non-parameter kernel density estimate was presented on the basis of abundant study on algorithms of moving object detection.
然后将总体框架转换为图像SR重建模型和运动估计模型；
The framework then can turn to SR resolution and motion estimation model.
将自适应迭代松弛搜索算法RSA引入到MPEG4模型基编码的三维运动参数估计中 ，解决了估计算法的稳健性问题 。
The problem of robustness in the estimation algorithm is solved by applying relax iterative search algorithm (RSA) to motion estimation of MPEG4 model based coding.
实证结果显示，模型较好地模拟了可转债实际价格运动路径，价格估计误差在5%以下。
The results show that the theoretical price of PLS model has a good fit with the actual price of convertible bond. The ratio of price error is lower than 5%.
实证结果显示，模型较好地模拟了可转债实际价格运动路径，价格估计误差在5%以下。
The results show that the theoretical price of PLS model has a good fit with the actual price of convertible bond. The ratio of price error is lower than 5%.
应用推荐