Hi,
does anybody use the area calculator plugin or the measure area fraction macro of the ImageJ? my problem is when I use any of these two applications the measured values in the TOTAL column are different, however I did not change anything (ROI or area ect.) thanks. |
On Aug 14, 2007, at 9:53 AM, Gergely Turi wrote:
> Hi, > does anybody use the area calculator plugin or the measure area > fraction > macro of the ImageJ? my problem is when I use any of these two > applications the > measured values in the TOTAL column are different, however I did not > change > anything (ROI or area ect.) > thanks. The Area Calculator plugin and MeasureAreaFraction macro are both obsolete. You can use the Measure command to measure ROI area and area fraction by checking "Area" and "Area Fraction" in the Analyze>Set Measurements dialog box. -wayne |
hello,
I would like to use ImageJ's analysis routines and plugins on massive multiresolution images. However, there is no way to work with these images in ImageJ, though there are simple Javascript/dHTML routines for working with them on the web (for example, see http://brainmaps.org/ajax-viewer.html?path=http://brainmaps.org/HBP2/c.aethiops/AGM1/AGM1-highres/350/&height=78181&width=92160 ) . I was wondering, is it possible to code an ImageJ plugin that will enable the importation of these types of massive multiresolution images, so that for example, they could be partially loaded into ImageJ and analysis routines run on portions of the images? It's inefficient to have to rely on taking screenshots of the web-based viewer and then importing into ImageJ, and would be much better if these massive multiresolution images could be opened directly and interactively in ImageJ, and analysis routines and plugins subsequently applied. Does anyone have any ideas on whether this is possible, or how to go about doing it? I have talked over this idea with Java programmers (which I am not) and they have not had any success implementing this idea, and so I'm wondering, is this a limitation of ImageJ or Java, or is it possible to create a plugin for loading massive multiresolution images directly into ImageJ, so that the user could interactively move to the region of interest within the massive image (as shown in the link above) and apply ImageJ's analysis routines? thanks. Shawn -- Shawn Mikula, Ph.D., Postdoctoral Scholar Center for Neuroscience University of California-Davis, 1544 Newton Court, Davis, CA 95618, Phone: 530-754-9209 Fax: 530-754-9136 mail: [hidden email] web: http://brainmaps.org |
Hi,
On Tue, 14 Aug 2007, Shawn Mikula wrote: > I would like to use ImageJ's analysis routines and plugins on massive > multiresolution images. However, there is no way to work with these > images in ImageJ, though there are simple Javascript/dHTML routines for > working with them on the web (for example, see > http://brainmaps.org/ajax-viewer.html?path=http://brainmaps.org/HBP2/c.aethiops/AGM1/AGM1-highres/350/&height=78181&width=92160 > ) . I was wondering, is it possible to code an ImageJ plugin that will > enable the importation of these types of massive multiresolution images, > so that for example, they could be partially loaded into ImageJ and > analysis routines run on portions of the images? Seems you are looking for TrakEM2 (see Plugins page on ImageJ's home page). Hth, Dscho |
thanks Dscho. I've looked into TrakEM2 but it seems that this is more
geared towards montaging applications using EM images. Also the idea of storing image tile locations in a Postgres db becomes unreasonable if you have millions of image tiles to deal with, and would be inefficient since the address for image tile locations is based on coordinates within the multiresolution image and can be computed using a fairly simple routine (rendering db storage unnecessary). However, I have been discussing this problem over with TrakEM2's creator, Albert Cardona, to see whether and what can be done for working with massive multiresolution images in ImageJ. thanks again. Shawn ----- Original Message ----- From: "Johannes Schindelin" <[hidden email]> To: "Shawn Mikula" <[hidden email]> Cc: <[hidden email]> Sent: Wednesday, August 15, 2007 7:29 AM Subject: Re: Partially Loading Massive Multiresolution Images Into ImageJ > Hi, > > On Tue, 14 Aug 2007, Shawn Mikula wrote: > >> I would like to use ImageJ's analysis routines and plugins on massive >> multiresolution images. However, there is no way to work with these >> images in ImageJ, though there are simple Javascript/dHTML routines for >> working with them on the web (for example, see >> http://brainmaps.org/ajax-viewer.html?path=http://brainmaps.org/HBP2/c.aethiops/AGM1/AGM1-highres/350/&height=78181&width=92160 >> ) . I was wondering, is it possible to code an ImageJ plugin that will >> enable the importation of these types of massive multiresolution images, >> so that for example, they could be partially loaded into ImageJ and >> analysis routines run on portions of the images? > > Seems you are looking for TrakEM2 (see Plugins page on ImageJ's home > page). > > Hth, > Dscho |
In reply to this post by Shawn Mikula-2
Thanks Jan. Your method is certainly useful in specific circumstances, and
yes, with 64-bit systems, you can have up to 16 exabytes (2^64 bits) of RAM, in principle. However, most people do not have over 2 GB RAM and what I'm trying to do is partially load a 30 GB image, which would make it impossible to load into RAM and would require just loading parts of it, preferably in an interactive manner so that you could zoom in and pan to a region of interest and then apply ImageJ's analysis routines to that portion of the image. What I've been doing is cropping these large images and importing the cropped portions into ImageJ for analysis (or alternatively, relying on screenshots using the online multiresolution image viewer); however, this is incredibly tedious and inefficient compared to the possibility of partially loading the multiresolution image into ImageJ and interactively going to the region of interest within the image for subsequent image analysis, all within ImageJ. I'm afraid in my previous email, I may not have made the problem completely clear. The problem is that the images I want to load into ImageJ exceed RAM, so they cannot be completely loaded. On the web, there is a straightforward way to view such large images using javascript/dHTML to grab just the visible image tiles of the large image, and examples of this can be found at Google Maps and brainmaps.org. What I would like to do is port this online javascript/dHTML viewer to ImageJ as a plugin, so that it there is the benefit of being able to rapidly apply ImageJ's analysis routines to visible portions of these massive images. Assuming you know the format of the massive multiresolution image (i.e., how to obtain image tile url's as a function of multiresolution image location in terms of x, y, and tier (i.e., level within the image pyramid)), then three variables are all that's needed for loading one of these massive images: 1) the image path, which can be a relative or absolute path, 2) the image height, and 3) the image width. As an ImageJ plugin, the user would have to specifiy these 3 variables before being able to load a massive multiresolution image. thanks again. Shawn ----- Original Message ----- From: "Jan de Sonneville" <[hidden email]> To: <[hidden email]> Sent: Tuesday, August 14, 2007 11:08 PM Subject: Re: Partially Loading Massive Multiresolution Images Into ImageJ > hello Shawn, > > I've found no problems loading in this amount of data, except that you > have to have enough memory to accomodate this. So this means that you > should look for the following requirements in a computer: > min 12 GB of ram and > a 64 bits operating system (windows 64 bit or linux 64 bit > http://www.ubuntu.com/getubuntu/download ) > > And in my previous post I've shown how to load the image in imageJ: > > " > java -jar -XmxNg -XX:+AggressiveHeap ij.jar > where you can replace N with the amount of gigabytes of ram you'd like to > use > > If you have multiple processors you can also add: > -XX:ParallelThreads=N --XX:+UseParallelOldGC > where N has to be replaced by the number of processors cores you'd like to > use (see dmesg) > > I've tested this using Sun jre version 6 (installable through package > manager for ubuntu 7.04, be sure that you use the sun version of java and > not the standard GPL one) > " > > Another possibility is to look at the plugin TrakEM2 by Albert Cardona > > Good luck, > Jan > > Date: Tue, 14 Aug 2007 10:37:09 -0700 > From: Shawn Mikula <[hidden email]> > Subject: Partially Loading Massive Multiresolution Images Into ImageJ > > hello, > I would like to use ImageJ's analysis routines and plugins on massive > multiresolution images. However, there is no way to work with these > images > in ImageJ, though there are simple Javascript/dHTML routines for working > with them on the web (for example, see > http://brainmaps.org/ajax-viewer.html?path=http://brainmaps.org/HBP2/c.aethiops/AGM1/AGM1-highres/350/&height=78181&width=92160 > ) > . I was wondering, is it possible to code an ImageJ plugin that will > enable > the importation of these types of massive multiresolution images, so that > for example, they could be partially loaded into ImageJ and analysis > routines run on portions of the images? It's inefficient to have to rely > on > taking screenshots of the web-based viewer and then importing into ImageJ, > and would be much better if these massive multiresolution images could be > opened directly and interactively in ImageJ, and analysis routines and > plugins subsequently applied. Does anyone have any ideas on whether this > is > possible, or how to go about doing it? I have talked over this idea with > Java programmers (which I am not) and they have not had any success > implementing this idea, and so I'm wondering, is this a limitation of > ImageJ > or Java, or is it possible to create a plugin for loading massive > multiresolution images directly into ImageJ, so that the user could > interactively move to the region of interest within the massive image (as > shown in the link above) and apply ImageJ's analysis routines? > > thanks. > Shawn > > > -- > Shawn Mikula, Ph.D., > Postdoctoral Scholar > Center for Neuroscience > University of California-Davis, > 1544 Newton Court, > Davis, CA 95618, > Phone: 530-754-9209 > Fax: 530-754-9136 > mail: [hidden email] > web: http://brainmaps.org > |
In reply to this post by Wayne Rasband
Thanks Wayne,
Unfortunately I forgot to mention that I set a threshold before the area measuring and for some reason I thought the Area value bears relation to the thresholded part of the ROI.
|
> Thanks Wayne,
> Unfortunately I forgot to mention that I set a threshold before the > area > measuring and for some reason I thought the Area value bears relation > to the > thresholded part of the ROI. The Measure command includes only the thresholded part of the selection when "Limit to Threshold" is enabled in Analyze>Set Measurements. -wayne |
The Measure command includes only the thresholded part of the selection
when "Limit to Threshold" is enabled in Analyze>Set Measurements. -wayne How can I get not only the size of the thresholded area but the whole size of the ROI in a thresholded picture? Is it the only way to calculate from the area fraction value? G. |
Hi Gergely,
Without "Limit to Threshold", all measurements are for the selection, thus "Area" gives the size of the ROI. Thus, you can deselect "Limit to Threshold" and do another measurement to get the area of the selection. Michael ________________________________________________________________ On 16 Aug 2007, at 16:39, Gergely Turi wrote: > The Measure command includes only the thresholded part of the > selection > when "Limit to Threshold" is enabled in Analyze>Set Measurements. > > -wayne > > How can I get not only the size of the thresholded area but the > whole size > of the ROI in a thresholded picture? Is it the only way to > calculate from > the area fraction value? > > G. |
Free forum by Nabble | Edit this page |