Putting index on centroids determined from image processing
显示 更早的评论
How can index each centroid determined from blob analysis?
hBlobAnalysis = vision.BlobAnalysis('MinimumBlobArea', 10000,'MaximumBlobArea',150000);
[objArea,objCentroid,bboxOut] = step(hBlobAnalysis, IBWopen);
So I have this part of my code that basically gets the centroid coordinates, what I would like to do is assign an index to each centroid determined and use it for an interative code later on.
回答（1 个）
Image Analyst
2022-3-27
You can do it this way
binaryImage = bwareafilt(binaryImage, [10000, 150000]);
props = regionprops(binaryImage, 'Centroid', 'BoundingBox');
xy = vertcat(props.Centroid);
imshow(binaryImage); % Display image.
hold on;
% Show boxes and centroids.
for k = 1 : length(props)
rectangle('Position', props(k).BoundingBox, 'EdgeColor', 'r');
xt = xy(k, 1);
yt = xy(k, 2);
str = sprintf(' (%.1f, %.1f)', xt, yt);
text(xt, yt, str, 'HorizontalAlignment', 'left', 'VerticalAlignment', 'middle')
plot(xt, yt, 'r+', 'MarkerSize', 30, 'LineWidth', 2)
end
4 个评论
Sean Ivan Roxas
2022-3-28
Thanks for responding, but I think you did not get exactly what I am trying to do. Attached above is the resulting image from the blob detection, as you can see on the right image, I have created a network wherein the nodes are the centroids of the coins and edges are created from the distance between the coins and its diameter (you actually helped me create this part as well). What I am trying to do now is to index each coin starting from the leftmost coin (index 1) to the rightmost coin (index N). Then I would create a vector that would contain the voltages of each coin (V_1,...,V_N) and another vector that would contain the current (I_1,...,I_N) such that it would update iteratively starting from the left coin then to its connected coins and so on until it reaches the rightmost coin. Then finally I will solve for the effective resistance via R = V/I.
Image Analyst
2022-3-28
Well of course I did not have your original image because you forgot to attach it. You said "gets the centroid coordinates, what I would like to do is assign an index to each centroid" so that is what I did. If your blobs are touching each other then you need to call imerode() to separate them.
Sean Ivan Roxas
2022-3-28
Image Analyst
2022-3-28
I don't usually use the Computer Vision toolbox way of finding blobs. Are you in a loop? Can't you just index the values
[objArea(k), objCentroid(k), bboxOut(k,:)] = step(hBlobAnalysis, IBWopen);
类别
在 帮助中心 和 File Exchange 中查找有关 Matrix Indexing 的更多信息
2022-3-27
2022-3-28
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
