A hardware design does calculate a normalised hog feature vector, which I want to visualise with the Matlab Hog library.
fpga_normArr = 'Output_Norm_FPGA.txt';
delimiterIn = '\n';
headerlinesIn = 0;
A = importdata(fpga_normArr,delimiterIn,headerlinesIn);
% want to give following function extractHOGFeatures "A" Vector as Input .. ?
img = imread('output_cropped.jpeg');
[_, hogVisualization] = extractHOGFeatures(fpga_normArr);
%override the hogVisualization output and plot the modified object?
plot(hogVisualization)
Is there a function in Matlab, which is able to plot a given feature vector?
Edit: I'm open for alternative approaches, perhabs on different platforms
Related
I am trying to fit a Poisson function to a histogram in Matlab: the example calls for using hist() (which is deprecated) so I want to use histogram() instead, especially as you cannot seem to normalize a hist(). I then want to apply a poisson function to it using poisspdf() or any other standard function (preferably no toolboxes!). The histogram is probability scaled, which is where the issue with the poisson function comes from AFAIK.
clear
clc
lambda = 5;
range = 1000;
rangeVec = 1:range;
randomData = poissrnd(lambda, 1, range);
histoFigure = histogram(randomData, 'Normalization', 'probability');
hold on
poissonFunction = poisspdf(randomData, lambda);
poissonFunction2 = poisspdf(histoFigure, lambda);
plot(poissonFunction)
plot(poissonFunction2)
I have tried multiple different methods of creating the poisson function + plotting and neither of them seems to work: the values within this function are not consistent with the histogram values as they differ by several decimals.
This is what the image should look like
however currently I can only get the bar graphs to show up correctly.
You're not specifing the x-data of you're curve. Then the sample number is used and since you have 1000 samples, you get the ugly plot. The x-data that you use is randomData. Using
plot(randomData, poissonFunction)
will lead to lines between different samples, because the samples follow each other randomly. To take each sample only once, you can use unique. It is important that the x and y values stay connected to each other, so it's best to put randomData and poissonFunction in 1 matrix, and then use unique:
d = [randomData;poissonFunction].'; % make 1000x2 matrix to find unique among rows
d = unique(d,'rows');
You can use d to plot the data.
Full code:
clear
clc
lambda = 5;
range = 1000;
rangeVec = 1:range;
randomData = poissrnd(lambda, 1, range);
histoFigure = histogram(randomData, 'Normalization', 'probability');
hold on
poissonFunction = poisspdf(randomData, lambda);
d = [randomData; poissonFunction].';
d = unique(d, 'rows');
plot(d(:,1), d(:,2))
With as result:
Does MatLab have any built in function to evaluate the density of a random variable from a custom histogram? (I suspect there are probably lots of ways to do this, I am just looking to see if there is already any builtin MatLab functionality).
Thanks.
The function hist gives you an approximation of the probability density you are evaluating.
If you want a continuous representation of it, this article from the Matlab documentation explains how to get one using the spline command from the Curve Fitting Toolbox. Basically the article explains how to make a cubic interpolation of your histogram.
The resulting code is :
y = randn(1,5001); % Replace y by your own dataset
[heights,centers] = hist(y);
hold on
n = length(centers);
w = centers(2)-centers(1);
t = linspace(centers(1)-w/2,centers(end)+w/2,n+1);
dt = diff(t);
Fvals = cumsum([0,heights.*dt]);
F = spline(t, [0, Fvals, 0]);
DF = fnder(F);
hold on
fnplt(DF, 'r', 2)
hold off
ylims = ylim;
ylim([0,ylims(2)]);
A popular way is to use kernel density estimation. The simplest way to do this in Matlab is using ksdensity.
I am doing a project to detect people in crowd using HOG-LBP. I want to make it for real-time application. I've read in some references, integral image/histogram can increase the speed of the performance from sliding window detection. I want to ask, how to implement integral image on my sliding window detection:
here is the code for integral image from matlab:
A = (cumsum(cumsum(double(img)),2));
and here my sliding window detection code:
im = strcat ('C:\Documents\Crowd_PETS09\S1\L1\Time_13-57\View_001\frame_0150.jpg');
im = imread (im);
figure (1), imshow(im);
win_size= [32, 32];
[lastRightCol lastRightRow d] = size(im);
counter = 1;
%% Scan the window by using sliding window object detection
% this for loop scan the entire image and extract features for each sliding window
% Loop on scales (based on size of the window)
for s=1
disp(strcat('s is',num2str(s)));
X=win_size(1)*s;
Y=win_size(2)*s;
for y = 1:X/4:lastRightCol-Y
for x = 1:Y/4:lastRightRow-X
%get four points for boxes
p1 = [x,y];
p2 = [x+(X-1), y+(Y-1)];
po = [p1; p2] ;
% cropped image based on the four points
crop_px = [po(1,1) po(2,1)];
crop_py = [po(1,2) po(2,2)];
topLeftRow = ceil(min(crop_px));
topLeftCol = ceil(min(crop_py));
bottomRightRow = ceil(max(crop_px));
bottomRightCol = ceil(max(crop_py));
cropedImage = im(topLeftCol:bottomRightCol,topLeftRow:bottomRightRow,:);
%Get the feature vector from croped image
HOGfeatureVector{counter}= getHOG(double(cropedImage));
LBPfeatureVector{counter}= getLBP(cropedImage);
LBPfeatureVector{counter}= LBPfeatureVector{counter}';
boxPoint{counter} = [x,y,X,Y];
counter = counter+1;
x = x+2;
end
end
end
where should i put the integral image code?
i am really appreciate, if someone can help me to figure it out.
Thank you.
The integral image is most suited for the Haar-like features. Using it for HOG or LBP would be tricky. I would suggest to first get your algorithm working, and then think about optimizing it.
By the way, the Computer Vision System Toolbox includes the extractHOGFeatures function, which would be helpful. Here's an example of training a HOG-SVM classifier to recognize hand-written digits. Also there is a vision.PeopleDetector object, which uses a HOG-SVM classifier to detect people. You could either use it directly for your project, or use it to evaluate performance of your own algorithm.
I have a large number (~1000) of files from a data logger that I am trying to process.
If I wanted to plot the trend from a single one of these log files I could do it using
plot(timevalues,datavalues)
I would like to be able to view all of these lines at same time in a similar way to how an oscilloscope has a "persistant" mode.
I can probably cobble together something that uses histograms but am hoping there is pre-existing or more elegant solution to this problem.
You can do exactly what you are suggesting yourself, i.e. plotting the heatmap of the signals.
Consider the following: I'll build a test signals (out of sine waves of different amplitude), then I'll plot the heatmap via hist3 and imagesc.
The idea is to build an auxiliary signal which is just the juxtaposition of all your time histories (both in x and y), then extract basic bivariate statistics out of that.
% # Test signals
xx = 0 : .01 : 2* pi;
center = 1;
eps_ = .2;
amps = linspace(center - eps_ , center + eps_ , 100 );
% # the auxiliary signal will be stored in the following variables
yy = [];
xx_f = [];
for A = amps
xx_f = [xx_f,xx];
yy = [yy A*sin(xx)];
end
% # final heat map
colormap(hot)
[N,C] = hist3([xx_f' yy'],[100 100]);
imagesc(C{1},C{2},N')
You can use also jet colormap instead of hot colormap for readability.
In the following the amplitude is gaussian instead of homogeneus.
here's a "primitive" solution that is just using hist:
%# generate some fake data
x=-8:0.01:8;
y=10*sinc(x);
yy=bsxfun(#plus,y,0.1*randn(numel(x),1000)' );
yy(randi(1000,1,200),:)= 5-randi(10)+ circshift(yy(randi(1000,1,200),:),[1 randi(numel(x),1,200)]);
%# get plot limit parameters
plot(x,yy)
yl=get(gca,'Ylim');
xl=get(gca,'Xlim');
close all;
%# set 2-d histogram ranges
ybins=100;
xbins=numel(x);
yrange=linspace(yl(1),yl(2),ybins);
xrange=linspace(xl(1),xl(2),xbins);
%# prealocate
m=zeros(numel(yrange),numel(xrange));
% build 2d hist
for n=1:numel(x)
ind=hist(yy(:,n),yrange);
m(:,n)=m(:,n)+ind(:);
end
imagesc(xrange,yrange,m)
set(gca,'Ydir','normal')
Why don't you normalize the data and then add all the lines together? You could then plot the heatmap from the single datafile.
I would like to use SURF (speed-up robust) features to implement a classification system based on a bag-of-visual-words approach. I have read some papers about the use of a dense grid to extract these features, but I can't find how to use it with MATLAB.
Does anybody know how to do SURF feature extraction on a dense grid with MATLAB?
[nRows, nCols] = size(grayImg);
STEP = 10;
colInd = (1 : STEP : nCols)';
rowInd = (1 : STEP : nRows)';
[A, B] = meshgrid(colInd, rowInd);
densePoints = [A(:) B(:)];
[featuresDense, validPointsDense] = extractFeatures(grayImg, densePoints, 'Method', 'SURF');
figure, imshow(img)
title('Dense SURF')
hold on
plot(validPointsDense)