MATLAB: Image tracking

centroidMATLAB

Hey, i am tracking a image using centroid concept. When i am calculating the centroid and standard deviation,it shows same value in many places and made bounding box in many place. how can i get rid off them to track the actual human in image.

Best Answer

I'm not sure why you show the same centroid in many places - doesn't seem right. About all I can suggest is that you check out sections 16 and 8 here: http://iris.usc.edu/Vision-Notes/bibliography/contents.html and do a search on "tracking" on MATLAB Central.

Related Solutions

MATLAB: Total area of a region of binary image

Huh? Of course you can use regionprops. No need to use bwboundaries() at all, unless you just want to show the outlines of the segmented areas in the overlay above your original gray scale or color image.

If you want the area of all the white pixels in your binary image you can do it two ways, with slightly different results because they use different calculation methods

pixelSum1 = bwarea(binaryImage);
pixelSum2 = sum(binaryImage(:));

If you want the area of each blob in the image individually, you can do it like this:

labeledImage = bwlabel(binaryImage);
measurements = regionprops(labeledImage, 'Area');
allAreas = [measurements.Area];  % List of all the blob areas.
totalAreaOfAllBlobs = sum(allAreas); % Will be the same as pixelSum2

MATLAB: Clearing border components from an imfreehand ROI

Thanks for the help. I dug a bit more and found a trick to get what I wanted. This is what I have used:

% Build a location array for the pixels on the edge of the freehand curve
% lassoxy.i and lassoxy.j are the index positions of the freehand curve.
% Img is the binary image produced by masking the original image by imfreehand.
locationxy=(lassoxy.j-1)*size(Img,1)+lassoxy.i;
% Use flood-fill operation to remove blobs touching the boundary (inside) of the freehand curve (negative binary image)
BW2 = imfill(~Img,locationxy);
Img_Bord = ~BW2; % Img_Bord is the binary image with all blobs touching the ROI edge removed.

Best Answer

Related Solutions

MATLAB: Total area of a region of binary image

MATLAB: Clearing border components from an imfreehand ROI

Related Question