Calculating perimeter of object!

Hi Folks,

So, I want to calculate the perimeter of an object. I have posted an image to this question. I wrote a short script - which I found on Mathworks - to trace the object of interest with a green trace. I was wondering if I can calculate the perimeter of this object. The perimeter measurement would be acquired first in number of pixels, and then I would convert that number into a physical SI measurement using a value I have written down.

My question is... How would I go about this? Also, would tracing the perimeter of the object, as I have done, help? Or, does tracing just serve to be a visual aid?

Here is the figure. The green is obviously the trace of the perimeter. In advance, thank you for reading through this!

Answer 1

[Walter Roberson]

Go back to the mask without the green traced in, and use regionprops() and ask for Perimeter

[Sean Farrell]

Okay, thanks! I just took a look at the function in MatLab, and it seems pretty sweet. How about if I am dealing with this image, shown below? How would I account for numerous objects? Pay no mind to the green trace; that was just me fiddling.

[Walter Roberson]

regionprops can handle multiple objects. But you need to clarify whether the areas that are thinly connected should be treated as one blob or should have their bounds broken and treated as multiple blobs.

[Sean Farrell]

Okay, thank you sir. I might come back to this thread to drop a quick question if I can't figure out how to apply the function to multiple objects. Thanks again!

[Sean Farrell]

If I have multiple objects, as in the image shown above, how do I know what perimeter belongs to which object?

[Image Analyst]

regionprops can handle multiple regions by returning a structure array:

props = regionprops(labeledImage, 'Area', 'Perimeter');
allAreas = [props.Area];
allPerimeters = [props.Perimeter];

To get the outlines as an array of (x,y) coordinates, use bwboundaries() and then use plot() to plot them all over the original image, like this snippet:

% bwboundaries() returns a cell array, where each cell contains the row/column coordinates for an object in the image.
% Plot the borders of all the regions on the original grayscale image using the coordinates returned by bwboundaries.
imshow(originalImage);
title('Outlines, from bwboundaries()', 'FontSize', captionFontSize); 
axis image; % Make sure image is not artificially stretched because of screen's aspect ratio.
hold on;
boundaries = bwboundaries(binaryImage);
numberOfBoundaries = size(boundaries, 1);
for k = 1 : numberOfBoundaries
  thisBoundary = boundaries{k};
  plot(thisBoundary(:,2), thisBoundary(:,1), 'g', 'LineWidth', 2);
end
hold off;

Answer 2

[nikhil agrawal]

Go back to the mask without the green traced in and use regionprops() and Ask for Perimeter.