2
$\begingroup$

My goal for this is to simply count the number of eggshells in the image.

eggshells

Using color negate only leads to this

shell = Import["https://i.stack.imgur.com/xJLJX.jpg"];
binshell = FillingTransform[ColorNegate[Binarize[shell]]]

negate

With luck, I at least came up with something clearer using

binshell = LocalAdaptiveBinarize[shell, 25];

binarize

What is the best way to count the total number of eggshells?

$\endgroup$
8
  • $\begingroup$ Are these photos always white on a green background? That would help separating items a lot. $\endgroup$
    – kirma
    Apr 28, 2018 at 18:14
  • $\begingroup$ Yep. I specifically chose green background. $\endgroup$
    – visatete
    Apr 28, 2018 at 18:16
  • $\begingroup$ Try ColorQuantize with parameter 2 (or maybe 3 or 4). $\endgroup$ Apr 28, 2018 at 18:18
  • $\begingroup$ Tried ColorQuantize. Using colornegate after only makes the whole image white. I however tried detecting all the white stuff using this as guide link but this is merely the result link $\endgroup$
    – visatete
    Apr 28, 2018 at 18:36
  • 1
    $\begingroup$ Very ad hoc, with too much magic: Import["https://i.imgur.com/PZNt7JQ.jpg"] // ColorSeparate[#, "S"] & // Binarize[#, 0.25] & // ColorNegate // Dilation[#, 5] & // MorphologicalComponents // Max (* 11 *) $\endgroup$
    – kirma
    Apr 28, 2018 at 18:54

1 Answer 1

10
$\begingroup$

Here's one way:

{r, g, b} = ColorSeparate[Import["https://i.imgur.com/PZNt7JQ.jpg"]];
bw = MorphologicalBinarize[b, 0.5]

enter image description here

ComponentMeasurements[bw, #Count > 200 &] // Length
11
$\endgroup$
1
  • 2
    $\begingroup$ Or Max@MorphologicalComponents@SelectComponents[bw, "Area", # > 200 &] in pre-v11. $\endgroup$
    – corey979
    Apr 28, 2018 at 19:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.