搜索资源列表
filter-in-video-sequences
- 粒子滤波理论是近年来跟踪领域的热门研究课题。在该领域,传统的卡尔曼(Kalman)滤波器是非常经典的运动目标跟踪工具。然而经典亦有其弊端,卡尔曼滤波对于非线性及非高斯环境下的工作能力相当无力。为解决这一问题,本文提出了一种基于粒子滤波的目标跟踪方法。其核心为以粒子(一种随机样本,携带权值)来表示后验概率密度,从而得到基于物理模型的近似最优数值解,其优点在于能在追踪的过程中实现更高的精度和更快的收敛速度等。粒子滤波通过加权计算这些带有权重的随机样本来得到目标的近似的运动状态,因此对于非高斯和非线性
fiuming
- 鲁棒性好,性能优越,最大似然(ML)准则和最大后验概率(MAP)准则,仿真效果非常好,插值与拟合,解方程,数据分析,多抽样率信号处理,对信号进行频谱分析及滤波,合成孔径雷达(SAR)目标成像仿真,时间序列数据分析中的梅林变换工具。 - Robustness, superior performance, Maximum Likelihood (ML) criteria and maximum a posteriori (MAP) criterion, Simulation of the eff
jaoten
- esprit算法对有干扰的信号频率进行估计,鲁棒性好,性能优越,对信号进行频谱分析及滤波。- esprit algorithm signal frequency interference can be assessed Robustness, superior performance, The signal spectral analysis and filtering.
fingjei_V1.6
- 鲁棒性好,性能优越,滤波求和方式实现宽带波束形成,基于matlab平台实现。- Robustness, superior performance, Filtering summation way broadband beamforming, Based on matlab platform.
kun_we76
- 鲁棒性好,性能优越,滤波求和方式实现宽带波束形成,最小二乘回归分析算法。- Robustness, superior performance, Filtering summation way broadband beamforming, Least-squares regression analysis algorithm.
MGMM_Particle-Filter
- 本文分别实现了整体模板更新和选择性子模块更新方法,以适应运动目标的运动姿态变化以及运动背景变化,并将其分别与粒子滤波目标跟踪算法相结合,以提高跟踪的鲁棒性。-This thesis studies and implements a total target model updating method and a selected sub-model updating method, and then combines it with the particle filter algorithm f
Kalman filtering
- Kalman filtering,卡尔曼滤波,本人用于智能车平衡组的角度滤波,滤波效果可观,鲁棒性强,调试时可观察波形来调整参数。(Kalman, filtering, Calman filtering, I used for intelligent vehicle balance group angle filtering, filtering effect is considerable, robustness, debugging can observe the waveform to a
svdwm1
- 基于svd的音频水印,加噪、滤波检验鲁棒性(Audio watermarking based on SVD, check the robustness of denoising and filtering)
MATLB
- 选择两张图片,一张水印图,一张嵌入图,将水印图进行Arnold置乱算法将其置乱,嵌入到嵌入图中,形成数字零水印,选用白噪声、高斯低通滤波、压缩、剪切、旋转攻击测试。以此观察图像鲁棒性(Select two pictures, a watermark and an embedded graph. We will scramble the watermark image with Arnold scrambling algorithm and embed it into the embedded m
VMD
- VMD方法是在传统维纳滤波的基础上,提出的一种非递归自适应信号分解新方法。与EMD方法和LMD方法相比,VMD方法分解的信号,具有精度高、收敛快和鲁棒性好等特点,适用于处理滚动轴承故障信号。(The VMD method is a new non recursive adaptive signal decomposition method based on traditional Wiener filtering. Compared with the EMD method and the LMD