Posted by
Michael Schmid on
Mar 01, 2016; 10:23am
URL: http://imagej.273.s1.nabble.com/Question-Can-image-J-make-a-Pixel-value-brightness-chart-tp5015749p5015753.html
Hi Thomas,
two more ideas:
(1) Take the average of the stack; this should be only the stars. Then threshold them, create a selection, enlarge it slightly and delete (set to background value).
Maybe the satellite is already the brightest object, then the maximum intensity of all slices would be sufficient. You can also set a selection (e.g. polygon) to restrict the measurement to the path of the satellite.
(2) Otherwise, write a macro:
Get the position in the first and last slice (see the getCursorLoc function)
For each slice, get the linear interpolation of the two positions, set a circular selection there and measure the mean or maximum. You can also have two circles (concentric), measure the mean pixel value and area of both and use this for background subtraction to get the desired intensity I:
mean1 = I/area1 + background
mean2 = I/area2 + background, and thus
mean1 - mean2 = I(1/area1 - 1/area2), which finally gives you the intensity without background
I = (mean1 - mean2) / (1/area1 - 1/area2)
Of course, it won't work if one of the circles contains a background star.
Michael
________________________________________________________________
On Mar 1, 2016, at 11:08, Herbie wrote:
> Good day Thomas,
>
> yes, the video was quite helpful for getting an idea of the issue!
>
> And no, I've no experience with astro-images but...
>
> My impression is, that there are several non-moving objects that change their intensity over time which makes simple frame to frame subtraction worthless. Consequently the moving object, the satellite, should somehow be tracked. There are ImageJ-plugins that are made for this but I've never used them. However, there may be a simpler approach if the track is more or less known. You could select the track---i.e. make it an (elongated) ROI (region of interest)--and perform frame to frame subtraction restricted to this ROI. If the background is sufficiently stable, this approach isolates the object and it should be rather easy to determine its gray value in every frame.
>
> You may just try the approach using a few frames and if it works you may write a little ImageJ-macro to automate the procedure.
>
> HTH
>
> Herbie
>
> :::::::::::::::::::::::::::::::::::::::::::
> Am 01.03.16 um 01:51 schrieb Thomas Dorman:
>> On Monday, February 29, 2016 5:46 PM, Thomas Dorman
>> <
[hidden email]> wrote:
>>
>>
>> HerbieThank you for your reply. I have a video of the North Korean
>> KMS 3-2 satellite passing through the field of view of my camera. The
>> satellites magnitude is varying as it tumbles. Was wanting to create
>> a chart showing the brightness variation of the satellite as the
>> satellite crosses the field. We are able to break the video down in
>> individual frames if needed.We know Image J is used a lot for micro
>> imaging but some people are using it for Astronomy. Do not know if
>> any one is attempting what we are trying to do with Image J. If it
>> would help to see the video it can be found here in three version
>> where others and you can understand what we are trying to
>> do.
http://vid284.photobucket.com/albums/ll34/skyflash_14700/KMS-3%20final%20video%201A_zpsmguano8n.mp4>>
>>
> Thanks you once again for your reply and any help in advance.RegardsThomas Dorman
>>
>>
>> On Monday, February 29, 2016 4:35 PM, Herbie <
[hidden email]>
>> wrote:
>>
>>
>> Good evening Thomas,
>>
>> it would be quite helpful to know what you mean by "Pixel value
>> (brightness) chart". I'm in the field of image processing for about
>> 40 years but have never heard of this.
>>
>> Is it the collection of the (gray)values that actually are present in
>> an image or is it the (gray)value histogram of an image?
>>
>> For the rest of us, please be more specific.
>>
>> Best
>>
>> Herbie
>>
>> ::::::::::::::::::::::::::::::::::::::::::: Am 29.02.16 um 23:07
>> schrieb Thomas Dorman:
>>> Can image J create a Pixel value (brightness) chart from a set of
>>> images or a video? If so where can we find the instructions to do
>>> so?RegardsThomas Dorman
>>>
>>> -- ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>>>
>>
>> -- ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>>
>>
>>
>>
>>
>> -- ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>>
>>
>
> --
> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html