资源简介
高斯混合 概率假设密度 滤波器 目标跟踪matlab代码及对应论文。可以直接按照说明运行出结果。
代码片段和文件信息
%CalculatePerformanceMetric
%Last modified 21 November 2013
%Matlab code by Bryan Clarke b.clarke@acfr.usyd.edu.au
%This is an implementation of the Optimal Subpattern Assignment
%(OSPA) metric proposed by Schuhmacher et al in
%Schuhmacher D.; Ba-Tuong Vo; Ba-Ngu Vo “A Consistent Metric for Performance Evaluation of Multi-object Filters“ Signal Processing IEEE Transactions on vol.56 no.8 pp.34473457 Aug. 2008
%X is the estimated state in the form [ [x1; y1; vx1; vy1] [x2; y2; vx2;
%vy2] ...]
%Y is the ground truth in the form [ [x1; y1; vx1; vy1] [x2; y2; vx2;
%vy2] ...]
%This isn‘t actually important as we will swap the labels so that X
%is the label of the shorter vector and Y is the label of the longer.
%cutoff_c and order_p are parameters that control the metric calculation;
%cutoff is a saturation threshold order controls how punishing it is to
%larger errors versus smaller ones. See the paper by Schuhmacher et al to
%get a handle on these in more detail.
%NOTE: This implementation is still a work in progress and is a bit buggy. Use with caution.
%NOTE: We only use 2D OSPA for position; we don‘t use the velocities.
function ospa = CalculateOSPAMetric(X Y cutoff_c order_p)
m = size(X 2);%Length of vector X
n = size(Y 2);%Length of vector Y
alphas = cutoff_c * ones(1 n);%Initialise to cutoff overwrite if there is a shorter value
bestOMATCost = -1;
bestOMATDataAssoc_i = [];
%m (i.e. the length of X) needs to be less than or equal to n (the length of Y)
if(m > n)%Swap them if this is not the case. X and Y are just labels that can be applied to either vector; whichever one is estimate or truth is not important.
tmpX = X;
tmpm = m;
X = Y;
m = n;
Y = tmpX;
n = tmpm;
end
if(m > 0)%If there are other values we need to find the best data association.
%We calculate all potential combinations (ie. subsampling without
%replacement)
comboSize = m;
valuesSampled = 1:n;
allCombos = combnk(valuesSampled comboSize);
nCombos = size(allCombos 1);
%Then for each combination we calculate every permutation (i.e.
%different ways to order the numbers) to find all possible data
%associations
for i = 1:nCombos
thisCombo = allCombos(i:);%The combination we are using
allDataAssocs = perms(thisCombo);
nDataAssocs = size(allDataAssocs 1);
%We check all the data associations for this combination
for j = 1:nDataAssocs
thisDataAssoc = allDataAssocs(j:);%An ordered list of the indices of Y to match them against the values in X
thisY = Y(:thisDataAssoc);
thisOMATCost = CalculateOMATMetric(X thisY order_p);
if(bestOMATCost < 0) || (thisOMATCost < bestOMATCost) %If thi
属性 大小 日期 时间 名称
----------- --------- ---------- ----- ----
目录 0 2014-05-09 21:13 GM_PHD_Filter_v110\
目录 0 2014-05-09 21:13 GM_PHD_Filter_v110\GM_PHD_Filter_v110\
目录 0 2014-05-09 21:13 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\
文件 4536 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\CalculateOSPAMetric.m
文件 2085 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\Calculate_Jacobian_H.m
文件 178 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\ConvertPlusMinusPi.m
文件 8343 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\error_ellipse.m
文件 2297 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Construct_Update_Components.m
文件 6743 2014-01-07 23:33 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Create_Birth.m
文件 14901 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Initialise_Jacobians.m
文件 2371 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Predict_Birth.m
文件 1958 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Predict_Existing.m
文件 2806 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Simulate_Initialise.m
文件 4092 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Simulate_Measurements.m
文件 3186 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Simulate_Plot.m
文件 6252 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_EKF_PHD_Update.m
文件 793 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Calculate_Performance_Metric.m
文件 1331 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Construct_Update_Components.m
文件 6516 2014-01-07 23:33 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Create_Birth.m
文件 2474 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Estimate.m
文件 5477 2014-01-07 23:39 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Filter.m
文件 10547 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Initialisation.m
文件 2332 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Predict_Birth.m
文件 1764 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Predict_Existing.m
文件 3458 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Prune.m
文件 2650 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Simulate_Initialise.m
文件 2770 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Simulate_Measurements.m
文件 2739 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Simulate_Plot.m
文件 6119 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\GM_PHD_Update.m
文件 9410 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\Hungarian.m
文件 1977 2014-01-07 23:19 GM_PHD_Filter_v110\GM_PHD_Filter_v110\GM_PHD_Filter\ospa_dist.m
............此处省略29个文件信息
- 上一篇:高斯混合PHD滤波器扩展目标跟踪
- 下一篇:盲图像质量评价
相关资源
- 高斯混合PHD滤波器扩展目标跟踪
- 高斯混合模型提取背景matlabcode
- 基于高斯混合模型GMM的说话人识别实
- 机器学习高斯混合模型资料总结含m
- 高斯混合模型EM算法Matlab代码
- EM算法在高斯混合模型中的应用有ma
- 基于matlab上实现高斯混合模型说话人
- 高斯混合PHD滤波器平滑
- GMM-matlab
- 高斯混合模型matlab实现
- matlab椒盐高斯混合噪声滤波
- GMM的matlab实现集合
- 随机集目标跟踪开发包
- 高斯混合概论假设密度滤波MATLAB代码
- 高斯混合模型的后验概率代码
- 高斯混合模型GMM 及高斯混合回归MAT
- 高斯混合模型说话人识别
- GMMP SkinColor
- 高斯混合模型(GMM)
- GMM,HMM的语音识别,说话人识别源码
- GMM 本代码建立高斯混合模型(高斯多
- GM-PHDsmooth
- GM-PHD 内容包括高斯混合概率假设密度
- 高斯混合模型matlab
- 高斯混合模型matlab代码
- 二维高斯混合模型GMM图形化简单明了
- 高斯混合模型,算法
评论
共有 条评论