How can we extract objects from a greyscale image?

So, i got a grayscale image of a box, which contains some objects.. I need to identify if/ not the box is empty or not....
I tried using the otsu thresholding, but doesn't work properly. is it possible to convert all the other parts to black except the image of interest?

3 Kommentare

How do you know which part is of interest before you have done the thresholding ?
This is the image of bucket which is full. the spatula, etc are the part of interest.
I would recommend background subtraction, if you can get a picture of the empty bucket, especially if everything is taken from a fixed viewpoint (if the viewpoint is not fixed then you might need to do image registration before doing the subtraction.)

Melden Sie sich an, um zu kommentieren.

Antworten (1)

darova
darova am 8 Aug. 2019
What about bwareaopen()
I = imread('example_filled2.jpg');
I1 = I < 40;
I2 = bwareaopen(I1,3e3);
imshow(I2)
What i've got
img22.png

Kategorien

Gefragt:

am 8 Aug. 2019

Kommentiert:

am 8 Aug. 2019

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by