how to diffrentiate diffrent objet in an image

[becool_hem]

hi all
i m doing work on object detection,i done how to calculate area & perimeter in am image for single object but how to calculate the same for the number of object
in my code if more than one object is present so the program calculate all the area of this object
plz help me if ther any direct fun in MATLAb
thank you in advance

 

[auto_mitch]

Could you please upload an image as sample in order to help you?

 

[becool_hem]

hi
i m load both the images
i already calculate area & perimeter of single object, can u plz tell me how to calculate area of no of object

 

[davidgrm]

To get the area what about the 'brute force way'? Count all the white pixels in the image. If you know what size the image represents then you will know the area of the objects

 

[auto_mitch]

As davidgrm said you can calculate the pixels of the background using the following formula background pixels= size of image - size of objects. if you cant find the objects' size let me know to help you!
e.g. You image needs illumination correction in order to extract all the objects!!! Have you tried this before?

 

[becool_hem]

thank for ur reply
i m using the same formula but cannot diffrentiate the no of object.
how to diffrentiate these object using color can u tell me any method for this
my aim is to sorting the object in an image using colr or area
thank you in adnvance

 

[lordsathish]

There is something called blob analysis...
I'm not sure if there are algorithms available on the net...
But I have seen many applications using this effectively.
This Blob analysis would give you the number of particles, their area and numerous such data...

Thanks

 

[becool_hem]

hi satish
thank for ur reply
blob analysis is also usable for face detection, but how to crop the different object in an image,if i m taking the images from webcamera how to handle the illumination

thanks in advance

 

[auto_mitch]

Message is unavailable.

 

[asit]

Try OpenCV

 

[lordsathish]

hi satish
thank for ur reply
blob analysis is also usable for face detection, but how to crop the different object in an image,if i m taking the images from webcamera how to handle the illumination

thanks in advance

Hi...
Blob analysis could be used to perform your requirement.
A group of pixels organized into a structure is commonly called a blob.
One of its application is face detection.
If you google you will find sufficient info on blob analysis.

 

[becool_hem]

hi satish
thanks for ur kind reply
i read the blob analysis but i hav a problem for example i hav an image contain 5 mans nad i want to crop only face and diffrentiate it
so how i can use this algorithm
plz help me out


thanks in advance

 

[naresh850]

there is a direct function in Matlab is region labeling.

search it.

aslo function for calculating all the objects area, perimeter,shape,centroid.

dont worry just go to help and search the region label function

 

[ztanish]

use open source library of sourceforge.com

 

[becool_hem]

hi to all
thanks you all for ur kind reply
if i m taking images from video, then how to detect the feverable object in an image.
wat is the the process to detect object in realtime(video format)
plz help me
thanking you in adavance

 

[swetha.incore]

Hi,
There is a concept of connected components ,in Matlab bwlabel will give you the number of connected components.It needs a black and white image. This is for a still image.
https://blogs.mathworks.com/steve/category/connected-components/.

This is not the answer for your question but it might give you an idea of how to approach the problem.

You can try color clustering algorithm for a video.I suppose the basic concept is from connected component analysis.
-swetha