Can somebody please help me detect an object using Principal Component Analysis and track its motion simultaneously using and comparing the performances of recursive Bayesian estimators Kalman and Particle filter?

I'm unable to generate a running code which can detect and track the motion using PCA and Kalman and Particle respectively.

5 Commenti

What code have you come up with so far, and what error message did you encounter?
I worked on a simple object detection code but its not running. function[x,y]=_sub(i_aft,i_bef) %subtract image i_sub=i_bef-i_aft;
%convert to binary igray=rgb2gray(i_sub); level=graythresh(i_sub); ibw=im2bw(i_sub,level);
%compute centre of gravity ibw = imfill(ibw,'holes'); L=bwlabel(ibw); stats=regionprops(L,'Area','Centroid'); ar=[stats.Area]; s=[stats.Centroid]; armax=max(ar); mx=max(max(L)); for i=1:mx if stats(i).Area==armax centroid=stats(i).Centroid; end end x=centroid(1); y=centroid(2); x=round(x); y=round(y); [m n r]=size(i_aft); for k=1:m for l= 1:n if (k==y && l==x) for t= 0:4 i_aft(k-t,l-t,:)= 0; i_aft(k-t,l-t,3)=255; i_aft(k-t,l+t,:)= 0; i_aft(k-t,l+t,3)=255; end end end end imshow(i_aft);
vid=videoinput('winvideo',1); triggerconfig(vid,'manual'); set(vid,'FramesPerTrigger',1); set(vid,'TriggerRepeat',Inf); start(vid);
while(1) for j=1:3 trigger(vid); im=getdata(vid,1); if j==1 i_bef=im; end if j==3 i_aft=im; end end [x,y,u]=i_sub(i_aft,i_bef); end
the error message encountered is :
Error using objdet (line 3)
Not enough input arguments.
Please edit your submission to use the code style option to make your code more readable.
Some thoughts:
The error you listed happens when you call a function without the necessary number of arguments.
Functions and variables cannot share a name in MATLAB. You should either rename your function i_sub, save the .m file under the new name, and delete i_sub.m or change the name of the variable i_sub.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%function[x,y,ibw]=isub(i_aft,i_bef)
%Takes two images from two frames ,subtracts them and detect the position
%of object by threshholding and calculating centroid
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function[x,y,ibw]=isub(i_aft,i_bef)
%subtract image
i_sub=i_bef-i_aft;
%convert to binary
igray=rgb2gray(i_sub);
level=graythresh(i_sub);
ibw=im2bw(i_sub,level);
%compute centre of gravity
ibw = imfill(ibw,'holes');
L=bwlabel(ibw);
stats=regionprops(L,'Area','Centroid');
ar=[stats.Area];
s=[stats.Centroid];
armax=max(ar);
mx=max(max(L));
for i=1:mx
if stats(i).Area==armax
centroid=stats(i).Centroid;
end
end
x=centroid(1);
y=centroid(2);
x=round(x);
y=round(y);
[m n r]=size(i_aft);
for k=1:m
for l= 1:n
if (k==y && l==x)
for t= 0:4
i_aft(k-t,l-t,:)= 0;
i_aft(k-t,l-t,3)=255;
i_aft(k-t,l+t,:)= 0;
i_aft(k-t,l+t,3)=255;
end
end
end
end
imshow(i_aft);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%Take images in real time and send to Function isub for processing
%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
vid=videoinput('winvideo',1);
triggerconfig(vid,'manual');
set(vid,'FramesPerTrigger',1);
set(vid,'TriggerRepeat',Inf);
start(vid);
while(1)
for j=1:3
trigger(vid);
im=getdata(vid,1);
if j==1
i_bef=im;
end
if j==3
i_aft=im;
end
end
[x,y,u]=isub(i_aft,i_bef);
end

Accedi per commentare.

Risposte (0)

Richiesto:

il 6 Gen 2014

Commentato:

il 22 Gen 2014

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by