AForge.NET

  :: AForge.NET Framework :: Articles :: Forums ::

Object within images' similarity

The forum is to discuss topics related to different areas of image processing and computer vision.

Object within images' similarity

Postby Lord Ra » Thu Mar 01, 2012 8:44 pm

Hello Folks,

I am a newbie and I was wondering if you would be kind enough to tell me if what I am attempting is doable with Aforge.
I am programming a web application in c # and i wish to compare an object in an uploaded image to the images stored in a database and arrive to the closest match.

Example scenario
Suppose the user uploads a image of a mobile phone, crops the image to focus mainly on the mobile phone. Now I have loads of images in the database and I would like to use Aforge to compare the user's cropped image to find a matching object in the images that are stored in the database.

I am considering using the Exhaustive template, image similarity comparison but I am not certain if it would do the job.

Is there an approach more suited than the one I am planning.

Thanks in advance,

Ra
Lord Ra
 
Posts: 1
Joined: Thu Mar 01, 2012 8:26 pm

Re: Object within images' similarity

Postby andrew.kirillov » Fri Mar 02, 2012 9:38 am

Hello,

Lord Ra wrote:I am considering using the Exhaustive template, image similarity comparison but I am not certain if it would do the job.

It will not help. 1) It is too slow to be used in video processing or comparing images with a database; 2) It is too simple to solve a task like this.

Lord Ra wrote:Is there an approach more suited than the one I am planning.

There is nothing in AForge.NET for things like this. Well, Kohonen network can be used to some extent (see some info here). But there is no advanced template matching or anything like this. Check SURF - it may be useful.
With best regards,
Andrew


Interested in supporting AForge.NET Framework?
User avatar
andrew.kirillov
Site Admin, AForge.NET Developer
 
Posts: 2767
Joined: Fri Jan 23, 2009 9:12 am
Location: UK




Return to Image Processing and Computer Vision