Help: measure and compare activity in a video quantifiably?

[LailaHiggins]

I have videos of starfish walking on a glass surface, filmed from underneath, 8 bit grey-scale.  I am trying to study the movement of the tube feet, and I need advice (help!) on how to proceed.  
Would it be possible to compare the difference in activeness of tube feet in one region of the video with another region?  Ideally I’d like to make a relative scale of activity so that I can compare, but I don’t know how to measure it in a quantifiable way.  

I have performed a stack registration (using the plugin StackReg), so that the frames are aligned to keep the starfish stationary.  

A brief description of the animal:
The starfish are spiny and have five rays (arms).  On the underside of the rays are the tube feet, the locomotory organs of the starfish.  Tube feet are small, cylindrical, extensible appendages on the underside of the starfish terminating in a flattened disc, typically 1 mm in diameter.  One starfish has hundreds of tube feet.  It is important to note that the tube feet in general are the same colour as the starfish, and therefore present very little contrast.  Also a whole starfish does not fit in the field of view, and as the sequence progresses, it walks out of the field of view.  
I really appreciate any ideas you might have on how I’d go about this,
Kind regards,
Laila.

[Ed Simmons]

Hi Leila,

On 16/10/2013 17:53, LailaHiggins wrote:
> I have videos of starfish walking on a glass surface, filmed from underneath,
> 8 bit grey-scale.  I am trying to study the movement of the tube feet, and I
> need advice (help!) on how to proceed.
> Would it be possible to compare the difference in activeness of tube feet in
> one region of the video with another region?  Ideally I’d like to make a
> relative scale of activity so that I can compare, but I don’t know how to
> measure it in a quantifiable way.

If you compute the difference between each frame and the next in the
stack, you will obtain data about which pixels have changed. You could
then blur this motion map image stack to get areas of activity, these
are then easy to quantify in some way...
>
> I have performed a stack registration (using the plugin StackReg), so that
> the frames are aligned to keep the starfish stationary.
This should help your cause if all you are interested is the relative
activity in different areas under your starfish.

> A brief description of the animal:
> The starfish are spiny and have five rays (arms).  On the underside of the
> rays are the tube feet, the locomotory organs of the starfish.  Tube feet
> are small, cylindrical, extensible appendages on the underside of the
> starfish terminating in a flattened disc, typically 1 mm in diameter.  One
> starfish has hundreds of tube feet.  It is important to note that the tube
> feet in general are the same colour as the starfish, and therefore present
> very little contrast.  Also a whole starfish does not fit in the field of
> view, and as the sequence progresses, it walks out of the field of view.
> I really appreciate any ideas you might have on how I’d go about this,
> Kind regards,
> Laila.
>
>
>

It shouldn't be too  hard to make a simple macro to test this out. Start
simple, and see if yuo can generate a motion map image stack, this in
itself will show you a lot just by looking at the images that are produced.

Hope that helps,
Ed

--
Ed Simmons
[hidden email]
www.esimagingsolutions.com

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Anderson, Charles (DNR)]

Leila,

I like Ed's suggestion, yet I think reviewers will want to know what activity is measured by the changes in light intensity. If you could first segment attached disks from the rest of the image, or otherwise locate attached disks, then you can at least say you are measuring features related to attachment activity.  If you post a small image as a new question, this list might be able to suggest specific methods to find and quantify disks.  

Charles

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Ed Simmons
Sent: Wednesday, October 16, 2013 11:14 AM
To: [hidden email]
Subject: Re: Help: measure and compare activity in a video quantifiably?

Hi Leila,

On 16/10/2013 17:53, LailaHiggins wrote:
> I have videos of starfish walking on a glass surface, filmed from
> underneath,
> 8 bit grey-scale.  I am trying to study the movement of the tube feet,
> and I need advice (help!) on how to proceed.
> Would it be possible to compare the difference in activeness of tube
> feet in one region of the video with another region?  Ideally I’d like
> to make a relative scale of activity so that I can compare, but I
> don’t know how to measure it in a quantifiable way.

If you compute the difference between each frame and the next in the stack, you will obtain data about which pixels have changed. You could then blur this motion map image stack to get areas of activity, these are then easy to quantify in some way...
>
> I have performed a stack registration (using the plugin StackReg), so
> that the frames are aligned to keep the starfish stationary.
This should help your cause if all you are interested is the relative activity in different areas under your starfish.

> A brief description of the animal:
> The starfish are spiny and have five rays (arms).  On the underside of
> the rays are the tube feet, the locomotory organs of the starfish.  
> Tube feet are small, cylindrical, extensible appendages on the
> underside of the starfish terminating in a flattened disc, typically 1
> mm in diameter.  One starfish has hundreds of tube feet.  It is
> important to note that the tube feet in general are the same colour as
> the starfish, and therefore present very little contrast.  Also a
> whole starfish does not fit in the field of view, and as the sequence progresses, it walks out of the field of view.
> I really appreciate any ideas you might have on how I’d go about this,
> Kind regards, Laila.
>
>
>

It shouldn't be too  hard to make a simple macro to test this out. Start simple, and see if yuo can generate a motion map image stack, this in itself will show you a lot just by looking at the images that are produced.

Hope that helps,
Ed

--
Ed Simmons
[hidden email]
www.esimagingsolutions.com

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html


--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html