how can I get the disjoit regions

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

how can I get the disjoit regions

贾舒婷(玮)

hi, all:
    on the white background , for example, there are three disjoint red regions, how can I get the red region 1, red region 2 and red region 3 separately? anyone can help? Thank you very much !
 
 
shu ting

_________________________________________________________________
超炫人气榜给您所有偶像的最新资讯和排名,快来支持自己的偶像!
http://cnweb.search.live.com/xrank/results.aspx?q=%e5%91%a8%e6%9d%b0%e4%bc%a6&FORM=MSNH

[Non-text portions of this message have been removed]

Reply | Threaded
Open this post in threaded view
|

Re: how can I get the disjoit regions

Suthakar Suthakar
Hi,

  Did you try thresholding the Image and creating a mask of the object.
  So from the mask it will be easy to detect the objects.

  You also use image labelling to get different objects separately.


Regards
Suthakar
--- In [hidden email], ¼ÖÊææã¨ç⣩ <jiaweizzu@...> wrote:
>
>
> hi, all:
>     on the white background , for example, there are three disjoint
red regions, how can I get the red region 1, red region 2 and red
region 3 separately? anyone can help? Thank you very much !
>  
>  
> shu ting
>
> _________________________________________________________________
> ³¬ìÅÈËÆø°ñ¸øÄúËùÓÐżÏñµÄ×îÐÂ×ÊѶºÍÅÅÃû£¬¿ìÀ´Ö§³Ö×Ô¼ºµÄżÏñ£¡
>
http://cnweb.search.live.com/xrank/results.aspx?q=%e5%91%a8%e6%9d%b0%e4%bc%a6&FORM=MSNH
>
> [Non-text portions of this message have been removed]
>


Reply | Threaded
Open this post in threaded view
|

Re: Re: how can I get the disjoit regions

Phani Bhushan Tholeti
> on the white background , for example, there are three disjoint
red regions, how can I get the red region 1, red region 2 and red
region 3 separately? anyone can help? Thank you very much !

Try sequential labeling - opencv has no implementation i believe, but
am not sure

On Thu, Jan 1, 2009 at 15:37, skar_06 <[hidden email]> wrote:

>   Hi,
>
> Did you try thresholding the Image and creating a mask of the object.
> So from the mask it will be easy to detect the objects.
>
> You also use image labelling to get different objects separately.
>
> Regards
> Suthakar
>
> --- In [hidden email] <OpenCV%40yahoogroups.com>, ¼ÖÊææã¨çâ£(c)
> <jiaweizzu@...> wrote:
> >
> >
> > hi, all:
>
> >
> >
> > shu ting
> >
> > __________________________________________________________
> > ³¬ìÅÈËÆø°ñ¸øÄúËùÓÐżÏñµÄ×îÐÂ×ÊѶºÍÅÅÃû£¬¿ìÀ´Ö§³Ö×Ô¼ºµÄżÏñ£¡
> >
>
> http://cnweb.search.live.com/xrank/results.aspx?q=%e5%91%a8%e6%9d%b0%e4%bc%a6&FORM=MSNH
> >
> > [Non-text portions of this message have been removed]
> >
>
>  
>



--
Lots o' Luv,
Phani Bhushan

Let not your sense of morals prevent you from doing what is right - Isaac
Asimov (Salvor Hardin in Foundation and Empire)


[Non-text portions of this message have been removed]

Reply | Threaded
Open this post in threaded view
|

RE: Re: how can I get the disjoit regions

贾舒婷(玮)

hi, all:
 
    Thanks a lot for all the response. I have tried blob analysis code from Dave Grossman, and can detect the disjoit regions (red bounding boxes) as shown in the example attached(1_1.tif). But I  need to get the separate blobs, then I found that if two bounding boxes have overlapping area, I can only get the separate blobs like (1_1_10.tif), in one bounding box, there's a part of content of the other one. anyone can solve this problem please? Cheers!
 
Shu ting



To: [hidden email]: [hidden email]: Fri, 2 Jan 2009 08:29:06 +0530Subject: Re: [OpenCV] Re: how can I get the disjoit regions



> on the white background , for example, there are three disjointred regions, how can I get the red region 1, red region 2 and redregion 3 separately? anyone can help? Thank you very much !Try sequential labeling - opencv has no implementation i believe, butam not sureOn Thu, Jan 1, 2009 at 15:37, skar_06 <[hidden email]> wrote:> Hi,>> Did you try thresholding the Image and creating a mask of the object.> So from the mask it will be easy to detect the objects.>> You also use image labelling to get different objects separately.>> Regards> Suthakar>> --- In [hidden email] <OpenCV%40yahoogroups.com>, ¼ÖÊææã¨çâ£(c)> <jiaweizzu@...> wrote:> >> >> > hi, all:>> >> >> > shu ting> >> > __________________________________________________________> > ³¬ìÅÈËÆø°ñ¸øÄúËùÓÐżÏñµÄ×îÐÂ×ÊѶºÍÅÅÃû£¬¿ìÀ´Ö§³Ö×Ô¼ºµÄżÏñ£¡> >>> http://cnweb.search.live.com/xrank/results.aspx?q=%e5%91%a8%e6%9d%b0%e4%bc%a6&FORM=MSNH> >> > [Non-text portions of this message have been removed]> >>> >-- Lots o' Luv,Phani BhushanLet not your sense of morals prevent you from doing what is right - IsaacAsimov (Salvor Hardin in Foundation and Empire)[Non-text portions of this message have been removed]





_________________________________________________________________
新版客服页面提供最新MSN客服资讯,助您轻松解决问题。
http://help.cn.msn.com/Cs.html

[Non-text portions of this message have been removed]

Reply | Threaded
Open this post in threaded view
|

Re: Re: how can I get the disjoit regions

ufuk aydin
In reply to this post by Phani Bhushan Tholeti
First, you can get the mask by thresholding, than use cvFloodFill to fill (with RGB=0,0,0) the mask region by region till there is no region left. This function will give you one connected component each time. The algorithm may be as follows:
 
1) Apply threshold and get a mask for those red regions,
2) Find a starting point (max point on mask) by cvMinMaxLoc function,
3) Fill that region (starting by that max point) with "black"=0, using cvFloodFill function,
4)This will give you one region and its rectangle,
5) Find another max point to start, if that max value is grater than 0.5,  go to step 3.



UFUK'tan !...

--- On Thu, 1/1/09, Phani Bhushan Tholeti <[hidden email]> wrote:

From: Phani Bhushan Tholeti <[hidden email]>
Subject: Re: [OpenCV] Re: how can I get the disjoit regions
To: [hidden email]
Date: Thursday, January 1, 2009, 9:59 PM






> on the white background , for example, there are three disjoint
red regions, how can I get the red region 1, red region 2 and red
region 3 separately? anyone can help? Thank you very much !

Try sequential labeling - opencv has no implementation i believe, but
am not sure

On Thu, Jan 1, 2009 at 15:37, skar_06 <skar_06@yahoo. com> wrote:

> Hi,
>
> Did you try thresholding the Image and creating a mask of the object.
> So from the mask it will be easy to detect the objects.
>
> You also use image labelling to get different objects separately.
>
> Regards
> Suthakar
>
> --- In OpenCV@yahoogroups. com <OpenCV%40yahoogrou ps.com>, ¼ÖÊææã¨çâ£(c)
> <jiaweizzu@. ..> wrote:
> >
> >
> > hi, all:
>
> >
> >
> > shu ting
> >
> > ____________ _________ _________ _________ _________ _________ _
> > ³¬ìÅÈËÆø°ñ¸øÄúËùÓÐżÏñµÄ×îÐÂ×ÊѶºÍÅÅÃû£¬¿ìÀ´Ö§³Ö×Ô¼ºµÄżÏñ£¡
> >
>
> http://cnweb. search.live. com/xrank/ results.aspx? q=%e5%91% a8%e6%9d% b0%e4%bc% a6&FORM=MSNH
> >
> > [Non-text portions of this message have been removed]
> >
>
>
>

--
Lots o' Luv,
Phani Bhushan

Let not your sense of morals prevent you from doing what is right - Isaac
Asimov (Salvor Hardin in Foundation and Empire)

[Non-text portions of this message have been removed]

 














     

[Non-text portions of this message have been removed]