How can I get correct output from bwlabel function?

1 次查看（过去 30 天）

Alaa 2016-2-28

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/270488-how-can-i-get-correct-output-from-bwlabel-function

评论： Walter Roberson 2016-2-29

Hi,

I've a problem with the output of bwlabel () function. I know exactly what is the input and the output of this function, but I don't know why does it give me incorrect output?

The problem is it takes all the cells in my image as one object. In other words, it gives NunOfobjs=1, which is really strange.

I have the following code:

    binaryImage = (imread('12_SS.bmp'));% binary image of type logical
    OriginalImage = imread('12.bmp');% original image (graysacle) of type uint8
    % Label each nerve so we can make measurements of it
    [labeledImage,NunOfobjs] = bwlabel(binaryImage, 8);    
    coloredLabels = label2rgb (labeledImage, 'hsv', 'k', 'shuffle');  
    figure, imshow(labeledImage, []); title('Labeled Image, from bwlabel()');
    figure, imagesc(coloredLabels);

The original image:

The binary image:

2 个评论
显示无隐藏无

Alaa 2016-2-28

I've got the correct answer, I think it counts the white parts as connected components, of which there is only one. Try running bwlabel for the negative of my image (or ~binaryImage), to look for the now-black spots.

Walter Roberson 2016-2-29

How did you calculate the binary image? It looks like perhaps you used edge detection, which is finding the boundaries between cells rather than finding the contents of the cells. Labeling the negation of your binary image would be a solution to that.

请先登录，再进行评论。

请先登录，再回答此问题。