After carrying out texture segmentation techniques on an image and then applying a binary mask I am trying to create an if else statement that will label the figure as a fail if the pixel count for black pixels in binary mask is above 1?

I am relatively new to this so New ideas are welcome, another way around this may be to have a function that will scan through the binary image pixel by pixel and if it picks up black pixels then how will I create an if else statement to label or have a print function saying picture equals 'FAIL'??

Here is my code:

I = imread('CYMDK_ABS.jpg')   //read image in// 

%figure, imshow(I)            // display initial image//

E = entropyfilt(I)            // apply entropy filter//

Eim = mat2gray(E)            // scale it down to a double image//

imshow(Eim)                  // show updated image//

BW1 = im2bw(Eim, .3)         // apply binary mask with threshold value .3//

defect = true                // define variable defect//

count = 0                    // define counter //
if count>1
    [row col]=find(BW1==0)      // this function returns all pixels equal to 0//
    defect = true                
else defect = false
    count = count + 1
end 

%imshow(BW1)

Answer 1

[Image Analyst]

It's not a very useful binary image if it must be all completely white. But anyway, to detect if there is a black pixel somewhere in the image:

if any(~BW1(:))
  % At least one pixel is black.
end

[Cormac]

Using the threshold '.3' was the best way to highlight the cluster. I could then use the ~BW1 to reverse the pixels when viewing it so the cluster would appear as white pixels rather than black. However thank you for the response. I appreciate the help.

Out of curiosity does this function scan through each pixel in the image one by one??

[Image Analyst]

any() looks at the image and processes each column, like many MATLAB functions such as sum(), etc. To avoid getting an array of numbers (one for each column) I turn the 2D array into a 1D array by using (:). Then any() checks to see if there is any pixel in that 1D array that is non-zero. Inverting it will tell you if there is at least one pixel that is exactly zero.