How to identify object using shape?

I want find object in given image for example I have butterfly image with matrix size is 123X235 and I given image contain butterfly with different size but same shape (ex:123X235,24X54,235X345,10X32).It is possible to identify the butterfly image ?

Antworten (1)

Image Analyst
Image Analyst am 27 Dez. 2016

0 Stimmen

Yes. Use imresize() and then run your algorithm on it. Or make an algorithm smart enough to identify the butterfly without depending on the image matrix size being a certain size.

6 Kommentare

CHINNI VENKATA RAMESH
CHINNI VENKATA RAMESH am 27 Dez. 2016
Sir all different sizes of butterfly exist in one image.
Image Analyst
Image Analyst am 27 Dez. 2016
I could not have known that from your description and because you didn't post an image. I think this may be too difficult of a project for you if you want to identify each butterfly individually. You could look up the fields of CBIR and convolutional neural networks, but I think you'll be getting in over your head. I think you should try something simpler.
Here's an example where they found jets in an image using the Hausdorf distance. http://cgm.cs.mcgill.ca/~godfried/teaching/cg-projects/98/normand/main.html
CHINNI VENKATA RAMESH
CHINNI VENKATA RAMESH am 28 Dez. 2016
sir i uploading image
CHINNI VENKATA RAMESH
CHINNI VENKATA RAMESH am 28 Dez. 2016
CHINNI VENKATA RAMESH
CHINNI VENKATA RAMESH am 28 Dez. 2016
Image Analyst
Image Analyst am 28 Dez. 2016
Well for that it's really easy. Just convert to hsv color space and threshold on saturation and call bwareafilt() to extract only the butterflies. But then you'll show me a picture of a flock of monarch butterflies coating a tree and ask me to count them and of course that simple algorithm won't work for that image.
See color segmentation methods in my File Exchange http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862 , like this one on hue segmentation that you can adapt to be on saturation segmentation.

Melden Sie sich an, um zu kommentieren.

Kommentiert:

am 28 Dez. 2016

Community Treasure Hunt

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

Start Hunting!

Translated by