Got Questions? Get Answers.
Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Thread Subject:
depth of the damage

Subject: depth of the damage

From: Beho nashed

Date: 18 Jun, 2010 16:55:21

Message: 1 of 10

Hello,
I have a question, please.
If I have a small specimen , and I know the thickness of it. if I know the color of the specimen both sides (same) , can I estimate the depth of the damage, depending on its color.

Thanks

Subject: depth of the damage

From: Walter Roberson

Date: 18 Jun, 2010 17:51:55

Message: 2 of 10

Beho nashed wrote:

> I have a question, please. If I have a small specimen , and I know the
> thickness of it. if I know the color of the specimen both sides (same) ,
> can I estimate the depth of the damage, depending on its color.

Maybe. Are you assuming that the specimen is of uniform composition and thus
uniform translucency and that the amount of light reflected or absorbed is
then proportional to the thickness of the specimen? Are you using a reflection
(illumination is from the top) or transmission (illumination is from the
bottom) model?

If the specimen is not of uniform composition then the calculation gets more
difficult, especially as there would be refraction at all internal boundaries.

Subject: depth of the damage

From: Beho nashed

Date: 19 Jun, 2010 16:07:05

Message: 3 of 10

Walter Roberson <roberson@hushmail.com> wrote in message <hvgbrp$ehj$1@canopus.cc.umanitoba.ca>...
> Beho nashed wrote:
>
> > I have a question, please. If I have a small specimen , and I know the
> > thickness of it. if I know the color of the specimen both sides (same) ,
> > can I estimate the depth of the damage, depending on its color.
>
> Maybe. Are you assuming that the specimen is of uniform composition and thus
> uniform translucency and that the amount of light reflected or absorbed is
> then proportional to the thickness of the specimen? Are you using a reflection
> (illumination is from the top) or transmission (illumination is from the
> bottom) model?
>
> If the specimen is not of uniform composition then the calculation gets more
> difficult, especially as there would be refraction at all internal boundaries.

Hello Walter,
It's transmission ( from the bottom ), and it's a uniform specimen.
Thanks

Subject: depth of the damage

From: ImageAnalyst

Date: 19 Jun, 2010 16:59:54

Message: 4 of 10

Of course. I'm sure you already know that color is frequently used to
assess quality of an object. But it's up to you to specify the color
ranges that constitute "good" and "bad" samples. You can do either
pass/fail, or a continuous kind of quality measurement number (such as
a delta E). Can you post some images? Not that you need images - you
can often do color measurement with a colorimeter or spectrophotometer
which creates no image at all.

Subject: depth of the damage

From: ImageAnalyst

Date: 19 Jun, 2010 17:01:54

Message: 5 of 10

As far as measuring the depth from the color, your best bet may be to
create a depth step wedge calibration standard. You'd have known
depths and then measure the color for that depth. Now you have a
calibration curve that can be used to look up the depth of any color.

Subject: depth of the damage

From: Beho nashed

Date: 19 Jun, 2010 17:43:06

Message: 6 of 10

ImageAnalyst <imageanalyst@mailinator.com> wrote in message <f6e18adb-8528-42d1-bf0a-db347fddffd8@d8g2000yqf.googlegroups.com>...
> As far as measuring the depth from the color, your best bet may be to
> create a depth step wedge calibration standard. You'd have known
> depths and then measure the color for that depth. Now you have a
> calibration curve that can be used to look up the depth of any color.

The same pics for the damage ==> http://drop.io/rlv072i (Second one)

I mean if I can detect the color of the surface ( in RGB) and have the it changes to the color to the other surface, in the other side. and the region should be the thickness. I wanna be able to say this damage happends in 2-2.3 mm of the sample, something like that. But I am not sure how to do that.

Subject: depth of the damage

From: Walter Roberson

Date: 19 Jun, 2010 22:52:34

Message: 7 of 10

Beho nashed wrote:

> The same pics for the damage ==> http://drop.io/rlv072i (Second one)
>
> I mean if I can detect the color of the surface ( in RGB) and have the
> it changes to the color to the other surface, in the other side. and the
> region should be the thickness. I wanna be able to say this damage
> happends in 2-2.3 mm of the sample, something like that. But I am not
> sure how to do that.

I don't know what I am looking at in the second image, but it appears to
be non-uniformly illuminated, with the illumination strongest at the
left and weakest at the right -- the shadows are much sharper at the
left. If you do not have uniform illumination then unless you can figure
out how to correct for that problem then you are not going to be able to
do a meaningful data analysis.

Subject: depth of the damage

From: ImageAnalyst

Date: 20 Jun, 2010 00:55:50

Message: 8 of 10

Yes, it turns out this is the same image he was asking about in a
different post. I also mentioned there that the image was a mess and
that he should get a good image to start with. He outlined two
different areas in the mess that he wanted automatically found, but it
wasn't clear how a program could automatically decide where those
boundaries should be. It's such a mess that I'm not really going to
take time to tackle it - it would just take too much time.

Subject: depth of the damage

From: Beho nashed

Date: 20 Jun, 2010 03:17:05

Message: 9 of 10

ImageAnalyst <imageanalyst@mailinator.com> wrote in message <ed61e950-c283-4f42-82f8-9775cd1c5359@k39g2000yqb.googlegroups.com>...
> Yes, it turns out this is the same image he was asking about in a
> different post. I also mentioned there that the image was a mess and
> that he should get a good image to start with. He outlined two
> different areas in the mess that he wanted automatically found, but it
> wasn't clear how a program could automatically decide where those
> boundaries should be. It's such a mess that I'm not really going to
> take time to tackle it - it would just take too much time.

That's right Imageanalyst, because it's the same project, I needed help with. All what I need is a simple code, that I can give it input of a specific color, (in RGB) and it will show a figure of the color dilates to white, or to another color I pick. If you can help, it would appreciated. Thanks

Subject: depth of the damage

From: ImageAnalyst

Date: 20 Jun, 2010 04:05:46

Message: 10 of 10

But it doesn't look that simple. I don't think simple code can give
you what you want.

Tags for this Thread

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Contact us