Removing unwanted object using regionprops

I have used function below, how can I change the pixel inside the boundary which the value less than threshold(0.8) to be 0(black)? Means that, I want to remove all objects which value less than 0.80. In my picture there are 7 objects, only two objects that I want it to maintain which has value of 0.96 and 0.84. I have tried add several line of command but failed. here the image that I means: http://www.flickr.com/photos/64698236@N03/6006901788/in/photostream

here the command that i used for regionprops:

[B,L] = bwboundaries(image7,'noholes');

% Display the label matrix and draw each boundary imshow(label2rgb(L, @jet, [.5 .5 .5])) hold on for k = 1:length(B) boundary = B{k}; plot(boundary(:,2), boundary(:,1), 'w', 'LineWidth', 2) end

stats = regionprops(L,'Area','Centroid');

threshold = 0.80;

% loop over the boundaries for k = 1:length(B)

% obtain (X,Y) boundary coordinates corresponding to label 'k' boundary = B{k};

% compute a simple estimate of the object's perimeter delta_sq = diff(boundary).^2; perimeter = sum(sqrt(sum(delta_sq,2)));

% obtain the area calculation corresponding to label 'k' area = stats(k).Area;

% compute the roundness metric metric = 4*pi*area/perimeter^2;

% display the results metric_string = sprintf('%2.2f',metric);

% mark objects above the threshold with a black circle if metric > threshold centroid = stats(k).Centroid; plot(centroid(1),centroid(2),'ko'); end

if metric < threshold L(i,j)=0; end

text(boundary(1,2)-35,boundary(1,1)+13,metric_string,'Color','y',... 'FontSize',14,'FontWeight','bold');

end

Thank you in advance.

Answer 1

[Image Analyst]

You don't even need bwboundaries or regionprops to get rid of pixels with a value of 0.8 or less. Just try something like (untested):

mask = image7 > 0.8
mask = cast(mask, class(image7));
outputImage = mask .* image7;

[Fathin]

Thank you for helping me..

I am sorry.I have tried that function but it doesn't works.

I am thinking of writing command ;

if metric < threshold

L(i,j) = 0

end

I tried this also, but it doesn't work, maybe i do wrote wrong.

[Fathin]

I want to remove that after the bwboundaries or regionprops calculate the parameters first.

You can refer to my picture above to get what I mean.

Answer 2

[Image Analyst]

There is nothing in the image that I can see that relates to the arbitrary text labels stuck next to your colored blobs. Not area, not intensity, not location, nothing. If you labeled and pseudocolored your blobs, then you can't do that - you'll have to work with the original grayscale image if the 0.8 is some kind of intensity value. Otherwise if you're dealing with this colored image you'll have to do something real ad hoc, such as getting rid of blobs based on their exact color rather than using the value 0.8 at all. Same thing if you use the labeled image - you'll just have to use the known label numbers of those blobs you want to remove rather than any sort of automatic removal process based on the value 0.8.

[Fathin]

I only colour the images. nothing to do with colour. I have the similar images in bw. So, I want to calculate the perimeter and area to get the value of roundness. after that, I want to remove the object which has value of roundness less than 0.8.

do you mean than I can't remove the object by using loop?

what do you mean to use the known label numbers of those blobs/object? how can I use label numbers of those blob?

thank you sir

[Fathin]

this is my command, I rewrite it;

[B,L] = bwboundaries(image7,'noholes');

% Display the label matrix and draw each boundary

imshow(label2rgb(L, @jet, [.5 .5 .5]))

hold on

for k = 1:length(B)

boundary = B{k};

plot(boundary(:,2), boundary(:,1), 'w', 'LineWidth', 2)

end

stats = regionprops(L,'Area','Centroid');

X = 0.80;

% loop over the boundaries

for k = 1:length(B)

% obtain (X,Y) boundary coordinates corresponding to label 'k'

boundary = B{k};

% compute a simple estimate of the object's perimeter

delta_sq = diff(boundary).^2;

perimeter = sum(sqrt(sum(delta_sq,2)));

% obtain the area calculation corresponding to label 'k'

area = stats(k).Area;

% compute the roundness metric

metric = 4*pi*area/perimeter^2;

% display the results

metric_string = sprintf('%2.2f',metric);

% mark objects above the X value with a black circle

if metric > X

centroid = stats(k).Centroid;

plot(centroid(1),centroid(2),'ko');

end

%this is the part that I added, I have tried the simplest and also the complex one but nothing works. I intend to make all pixel in the object which has value less than X to turn black (originally white)

if metric < X

L(i,j)=0;

end

text(boundary(1,2)-35,boundary(1,1)+13,metric_string,'Color','y','FontSize',14,'FontWeight','bold');

end

[Harini Ramasamy]

how to calculate the number of similar metric value within some range for example if i had a many range of roundness value then how to calculte the number of similar range of roundness like from .2 to .8