problem with centroid of binary image

hi i use bwconncomp and regionprob with centroid feature to find centroid of binary image but i cant understand one thing? if i use centroid feature, it return one point or a structure of points? when i use this i have a structure of point that i think it is wrong do you have any idea?

[nadia naji]

i need to consider the behavior of centroid of binary image (with several object in that image ) in sequential frame of video and show them the number of object can be different in each frame. how can i do it? pls help me?

[Image Analyst]

See my BlobsDemo tutorial : http://www.mathworks.com/matlabcentral/fileexchange/25157-image-segmentation-tutorial-blobsdemo, or see my demo program below.

Answer 1

[Walter Roberson]

Note that if your "binary array" is a numeric datatype, with 0's and 1's, rather than a logical array (false and true's), then regionprops will treat the numeric array as if it is a labeled array that happens to contain only one label. To avoid that, call bwlabel() to separate out the blobls.

labelarray = bwlabel(not_really_binary_Image));
measurements = regionprops(labelarray, 'Centroid');

then measurements will be a struct array with the field 'Centroid' for each blob. Each of the 'Centroid' fields will be a vector whose length matches the dimension of the image (usually 2).

[Image Analyst]

It was that way for older versions (prior to R2008a I think??). With more modern versions it considers them as separate blobs if the values are all 1, though it will merge blobs for disconnected blobs with number 2 or higher. See this demo where I skip labeling in R2012b and it finds 3 separate blobs:

grayImage = peaks(200);
subplot(1,2,1);
imshow(grayImage, []);
title('Original Gray Image', 'FontSize', 30);
axis on;
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);
% Binarize the image to find 3 blobs.
binaryImage  = grayImage > 3;
subplot(1,2,2);
imshow(binaryImage);
axis on;
title('Binary Image with Marked Centroids', 'FontSize', 30);
% Make the measurements, skipping the labeling.
measurements = regionprops(binaryImage, 'Area', 'Centroid')
[allAreas] = [measurements.Area]
allCentroids = [measurements.Centroid]
x_centroids = allCentroids(1:2:end)
y_centroids = allCentroids(2:2:end)
hold on;
plot(x_centroids, y_centroids, 'r+', 'MarkerSize', 30);

[nadia naji]

is not any way to find the centroid of whole image in r2012a? can you help me?

[Walter Roberson]

The documentation indicates http://www.mathworks.com/help/images/ref/regionprops.html that the automatic labelling happens for binary arrays, which the documentation defines in terms of logical datatype

[nadia naji]

if i mean on structure of centroid it can get me the centroid of objects in each frame?

[Walter Roberson]

no, you cannot apply mean() to a structure.

what does class() of your binary array show? Are you passing the binary array into regionprops or are you passing the result of bwconncomp ? bwconncomp already does the separation into blobs so if you want the overall centroid rather than the centroid of each blob, you should not pass in the result of bwconncomp

[Image Analyst]

Walter - you're right. I didn't read carefully enough.

[nadia naji]

i use bwconncomp and then pass the output of this function to regionprobs in fact i use this code con=bwconncomp(img); ctr=regionprobs(con,'Centroid');

is it wrong? i should pass the image to regionprobs?? i pass the binary image to regionprobs but i have a structure of centroids instead of only one centroid for overall image?

[Walter Roberson]

If you pass the con to regionprops you are going to get one centroid per blob.

If you pass an array of logical values (true & false) to modern regionprops() then you are going to get one centroid per blob.

If you do have an array of logical values as your variable img, then

ctr = regionprops(double(img), 'Centroid');

should return the centroid of the image.

[Image Analyst]

Please post your image. I don't like working "blind."

[nadia naji]

thanks a lot it works by using double i really appreciate your help

[Image Analyst]

How did you end up with a double? That's not normally the case. Usually you process the image to a point where you can threshold it and get a binary image. You said you had a binary image (logical) but now you say you have a double image. Something's weird.

Answer 2

[Image Analyst]

regionprops returns a structure array. Each element in the array is a structure with all the measurements for that one blob. Centroid or WeightedCentroid might be one of those measurements, if you asked regionprops to calculate it. It will be a 2 element array. For example, for blob #k,

measurements = regionprops(binaryImage, 'Centriod');
theCentroid = measurements(k).Centroid; % for blob #k only

Perhaps this demo will help:

% Demo to have the user freehand draw an irregular shape over
% a gray scale image, have it extract only that part to a new image,
% and to calculate the mean intensity value of the image within that shape.
% Also calculates the perimeter, centroid, and center of mass (weighted centroid).

% Change the current folder to the folder of this m-file.
if(~isdeployed)
  cd(fileparts(which(mfilename)));
end
clc;  % Clear command window.
clear;  % Delete all variables.
close all;  % Close all figure windows except those created by imtool.
imtool close all;  % Close all figure windows created by imtool.
workspace;  % Make sure the workspace panel is showing.
fontSize = 16;

% Read in a standard MATLAB gray scale demo image.
folder = fullfile(matlabroot, '\toolbox\images\imdemos');
baseFileName = 'cameraman.tif';
% Get the full filename, with path prepended.
fullFileName = fullfile(folder, baseFileName);
% Check if file exists.
if ~exist(fullFileName, 'file')
  % File doesn't exist -- didn't find it there.  Check the search path for it.
  fullFileName = baseFileName; % No path this time.
  if ~exist(fullFileName, 'file')
    % Still didn't find it.  Alert user.
    errorMessage = sprintf('Error: %s does not exist in the search path folders.', fullFileName);
    uiwait(warndlg(errorMessage));
    return;
  end
end
grayImage = imread(fullFileName);
imshow(grayImage, []);
axis on;
title('Original Grayscale Image', 'FontSize', fontSize);
set(gcf, 'Position', get(0,'Screensize')); % Maximize figure.
message = sprintf('Left click and hold to begin drawing.\nSimply lift the mouse button to finish');
uiwait(msgbox(message));
hFH = imfreehand();
% Create a binary image ("mask") from the ROI object.
binaryImage = hFH.createMask();
xy = hFH.getPosition;

% Now make it smaller so we can show more images.
subplot(2, 3, 1);
imshow(grayImage, []);
axis on;
drawnow;
title('Original Grayscale Image', 'FontSize', fontSize);

% Display the freehand mask.
subplot(2, 3, 2);
imshow(binaryImage);
axis on;
title('Binary mask of the region', 'FontSize', fontSize);

% Label the binary image and computer the centroid and center of mass.
labeledImage = bwlabel(binaryImage);
measurements = regionprops(binaryImage, grayImage, ...
    'area', 'Centroid', 'WeightedCentroid', 'Perimeter');
area = measurements.Area
centroid = measurements.Centroid
centerOfMass = measurements.WeightedCentroid
perimeter = measurements.Perimeter

% Calculate the area, in pixels, that they drew.
numberOfPixels1 = sum(binaryImage(:))
% Another way to calculate it that takes fractional pixels into account.
numberOfPixels2 = bwarea(binaryImage)

% Get coordinates of the boundary of the freehand drawn region.
structBoundaries = bwboundaries(binaryImage);
xy=structBoundaries{1}; % Get n by 2 array of x,y coordinates.
x = xy(:, 2); % Columns.
y = xy(:, 1); % Rows.
subplot(2, 3, 1); % Plot over original image.
hold on; % Don't blow away the image.
plot(x, y, 'LineWidth', 2);
drawnow; % Force it to draw immediately.

% Burn line into image by setting it to 255 wherever the mask is true.
burnedImage = grayImage;
burnedImage(binaryImage) = 255;
% Display the image with the mask "burned in."
subplot(2, 3, 3);
imshow(burnedImage);
axis on;
caption = sprintf('New image with\nmask burned into image');
title(caption, 'FontSize', fontSize);

% Mask the image and display it.
% Will keep only the part of the image that's inside the mask, zero outside mask.
blackMaskedImage = grayImage;
blackMaskedImage(~binaryImage) = 0;
subplot(2, 3, 4);
imshow(blackMaskedImage);
axis on;
title('Masked Outside Region', 'FontSize', fontSize);
% Calculate the mean
meanGL = mean(blackMaskedImage(binaryImage));

% Put up crosses at the centriod and center of mass
hold on;
plot(centroid(1), centroid(2), 'r+', 'MarkerSize', 30, 'LineWidth', 2);
plot(centerOfMass(1), centerOfMass(2), 'g+', 'MarkerSize', 20, 'LineWidth', 2);

% Now do the same but blacken inside the region.
insideMasked = grayImage;
insideMasked(binaryImage) = 0;
subplot(2, 3, 5);
imshow(insideMasked);
axis on;
title('Masked Inside Region', 'FontSize', fontSize);

% Now crop the image.
leftColumn = min(x);
rightColumn = max(x);
topLine = min(y);
bottomLine = max(y);
width = rightColumn - leftColumn + 1;
height = bottomLine - topLine + 1;
croppedImage = imcrop(blackMaskedImage, [leftColumn, topLine, width, height]);
% Display cropped image.
subplot(2, 3, 6);
imshow(croppedImage);
axis on;
title('Cropped Image', 'FontSize', fontSize);

% Put up crosses at the centriod and center of mass
hold on;
plot(centroid(1)-leftColumn, centroid(2)-topLine, 'r+', 'MarkerSize', 30, 'LineWidth', 2);
plot(centerOfMass(1)-leftColumn, centerOfMass(2)-topLine, 'g+', 'MarkerSize', 20, 'LineWidth', 2);

% Report results.
message = sprintf('Mean value within drawn area = %.3f\nNumber of pixels = %d\nArea in pixels = %.2f\nperimeter = %.2f\nCentroid at (x,y) = (%.1f, %.1f)\nCenter of Mass at (x,y) = (%.1f, %.1f)\nRed crosshairs at centroid.\nGreen crosshairs at center of mass.', ...
meanGL, numberOfPixels1, numberOfPixels2, perimeter, ...
centroid(1), centroid(2), centerOfMass(1), centerOfMass(2));
msgbox(message);

[nadia naji]

excuse me i cant understand some things i need to have one centroid for objects in image not structure of several centroid because i need to consider the movement of centroids in sequential frame? do you understand what i need? i need centroid of whole image please help me

[Image Analyst]

Nadia, if this is in a loop over frames, just extract the centroids (x_centroids and y_centroids) from the structure. Make x_centroids and y_centroids arrays that depend on the frame. So if you have 300 frames, x_centroids is an array of 300 x centroids. Then you can do mean_x_centroid = mean(x_centroids). It becomes more complicated if there are different numbers of blobs in each frame - then it becomes essentially a tracking problem, which is more complicated because you have to account for blobs coming into the field of view and blobs leaving the field of view.