linear Hough transform
Locked
 
|
Hello All,
I need to detect faint linear signal (fine processes on neurons) in images. I wanted to try using a linear Hough transform. In the listserv archives a "Linear_Hough_Transform_.class plugin" is mentioned, but I can not find it. Can someone tell me where it is (the source code)? If the source code for this plugin is not available, I wondered if someone could tell me if it would be relatively easy to convert the Hough Circles plugin to detect lines or if that would be a bad way to go about things. Thanks, Martha Narro |
|
|
The principle of the Hough transform is the same for lines or circles or any shape. So it would be relatively easy to modify the circles transform. I however currently have no time to do so. Good luck
Michael D. Abràmoff, MD, PhD Assistant Professor of Clinical Ophthalmology / Assistant Professor of Electrical and Computer Engineering Department of Ophthalmology and Visual Sciences University of Iowa Hospital and Clinics, PFP 11290C 200 Hawkins Drive Iowa City, IA 52242 USA Tel: +1 319 384 5833. Secretary Val Bell: +1 319 356 3938. Skype michael-abramoff [hidden email] -----Original Message----- From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Martha Narro Sent: Friday, August 19, 2005 2:45 PM To: [hidden email] Subject: linear Hough transform Hello All, I need to detect faint linear signal (fine processes on neurons) in images. I wanted to try using a linear Hough transform. In the listserv archives a "Linear_Hough_Transform_.class plugin" is mentioned, but I can not find it. Can someone tell me where it is (the source code)? If the source code for this plugin is not available, I wondered if someone could tell me if it would be relatively easy to convert the Hough Circles plugin to detect lines or if that would be a bad way to go about things. Thanks, Martha Narro |
Re: linear Hough transform
|
|
In reply to this post by Martha Narro
Dear Martha,
Why don't you try conventional edge-detection with liner filters ( say Sobel)? It is already implemented in ImageJ. I am not sure how the Hough transform handles images with degraded geometric forms (e.g. noisy images). The other thing you can try is to threshold the images and then skeletonize them. best regards Dimiter Date: Fri, 19 Aug 2005 12:44:48 -0700 From: Martha Narro <[hidden email]> Subject: linear Hough transform Hello All, I need to detect faint linear signal (fine processes on neurons) in images. I wanted to try using a linear Hough transform. In the listserv archives a "Linear_Hough_Transform_.class plugin" is mentioned, but I can not find it. Can someone tell me where it is (the source code)? If the source code for this plugin is not available, I wondered if someone could tell me if it would be relatively easy to convert the Hough Circles plugin to detect lines or if that would be a bad way to go about things. Thanks, Martha Narro |
|
|
Dear Dimiter,
I'm also interested in the linear Hough Transform. By example, if I have four lines to detect in an image and I want their parameters (r, thêta), the Hough tranform gives me these parameters, no ? Is it possible with a Sobel operator to easily estimate the line parameters ? Best regards, Fleur Prodanov, D. (FYS) a écrit : >Dear Martha, > >Why don't you try conventional edge-detection with liner filters ( say Sobel)? It >is already implemented in ImageJ. I am not sure how the Hough transform handles images with >degraded geometric forms (e.g. noisy images). The other thing you can try is to >threshold the images and then skeletonize them. > >best regards > >Dimiter > >Date: Fri, 19 Aug 2005 12:44:48 -0700 >From: Martha Narro <[hidden email]> >Subject: linear Hough transform > >Hello All, > >I need to detect faint linear signal (fine processes on neurons) in images. >I wanted to try using a linear Hough transform. In the listserv archives a >"Linear_Hough_Transform_.class plugin" is mentioned, but I can not find it. >Can someone tell me where it is (the source code)? If the source code for >this plugin is not available, I wondered if someone could tell me if it >would be relatively easy to convert the Hough Circles plugin to detect >lines or if that would be a bad way to go about things. > >Thanks, >Martha Narro > > > > -- Fleur LONGUETAUD Centre INRA de Nancy EQUIPE QUALITE DES BOIS Laboratoire d'Etude des Ressources Forêt-Bois UMR INRA-ENGREF F-54280 CHAMPENOUX (France) http://www.nancy.inra.fr/lerfob/ |
|
|
In reply to this post by Martha Narro
Dear All,
Thanks for the help. Fleur & Philipp sent me the plugin. Now that I have it and see who the author is, I realize I could not get it with a Google search because, although it comes up in the search, the link no longer works. I will play with it. I will also look into Sobel operators as Dimiter suggested. I'll have to do some reading. I already know that Laplacian edge detection and skeletonization tend to miss the fine neuronal processes I am trying to detect. Thanks again, Martha Narro On Mon, 22 Aug 2005 11:04:55 +0200, Fleur Longuetaud <[hidden email]> wrote: >Dear Dimiter, > >I'm also interested in the linear Hough Transform. >By example, if I have four lines to detect in an image and I want their >parameters (r, thêta), the Hough tranform gives me these parameters, no ? >Is it possible with a Sobel operator to easily estimate the line >parameters ? >Best regards, > >Fleur > >Prodanov, D. (FYS) a écrit : > >>Dear Martha, >> >>Why don't you try conventional edge-detection with liner filters ( say >>is already implemented in ImageJ. I am not sure how the Hough transform handles images with >>degraded geometric forms (e.g. noisy images). The other thing you can try is to >>threshold the images and then skeletonize them. >> >>best regards >> >>Dimiter >> >>Date: Fri, 19 Aug 2005 12:44:48 -0700 >>From: Martha Narro <[hidden email]> >>Subject: linear Hough transform >> >>Hello All, >> >>I need to detect faint linear signal (fine processes on neurons) in images. >>I wanted to try using a linear Hough transform. In the listserv archives a >>"Linear_Hough_Transform_.class plugin" is mentioned, but I can not find it. >>Can someone tell me where it is (the source code)? If the source code for >>this plugin is not available, I wondered if someone could tell me if it >>would be relatively easy to convert the Hough Circles plugin to detect >>lines or if that would be a bad way to go about things. >> >>Thanks, >>Martha Narro >> >> >> >> > > >-- >Fleur LONGUETAUD >Centre INRA de Nancy >EQUIPE QUALITE DES BOIS >Laboratoire d'Etude des Ressources Forêt-Bois >UMR INRA-ENGREF >F-54280 CHAMPENOUX (France) > >http://www.nancy.inra.fr/lerfob/ >========================================================================= |
|
|
Can anybody sent a the plugin please?
email: tigersmail@gmail.com Thanks a lot
|
«
Return to ImageJ
|
1 view|%1 views
| Free forum by Nabble | Edit this page |