| |
 |
|
| Science Forum Index » Image Processing Forum » Image gradient?... |
|
Page 1 of 1 |
|
| Author |
Message |
| mlt... |
 Posted: Wed Jul 01, 2009 4:20 am |
|
|
|
Guest
|
On this page its said that the gradient points in the direction of most
increase in intensity:
http://en.wikipedia.org/wiki/Image_gradient
but why are all the arrows pointing toward the black area? Is white not
defined as having the highest intensity? |
|
|
| Back to top |
|
|
|
| Martin Brown... |
| Posted: Wed Jul 01, 2009 4:33 am |
|
|
|
Guest
|
mlt wrote:
[quote:90e336fce4]On this page its said that the gradient points in the direction of most
increase in intensity:
http://en.wikipedia.org/wiki/Image_gradient
but why are all the arrows pointing toward the black area? Is white not
defined as having the highest intensity?
[/quote:90e336fce4]
And so the downhill direction is...
Regards,
Martin Brown |
|
|
| Back to top |
|
|
|
| mlt... |
| Posted: Wed Jul 01, 2009 4:49 am |
|
|
|
Guest
|
"Martin Brown" <|||newspam||| at (no spam) nezumi.demon.co.uk> wrote in message
news:yXG2m.295$3o6.107 at (no spam) newsfe24.iad...
[quote:a9a3780353]mlt wrote:
On this page its said that the gradient points in the direction of most
increase in intensity:
http://en.wikipedia.org/wiki/Image_gradient
but why are all the arrows pointing toward the black area? Is white not
defined as having the highest intensity?
And so the downhill direction is...
Regards,
Martin Brown
[/quote:a9a3780353]
From the page:
"At each image point, the gradient vector points in the direction of largest
possible intensity increase, and the length of the gradient vector
corresponds to the rate of change in that direction"
If white is max an black is 0 then the arrow should point toward the white
area right? Unless start is where the arrowhead is located, but that would
not make sense.
Not sure what you mean by downhill direction. |
|
|
| Back to top |
|
|
|
| Jonathan Campbell... |
| Posted: Wed Jul 01, 2009 6:35 am |
|
|
|
Guest
|
mlt wrote:
[quote:be21356240]"Martin Brown" <|||newspam||| at (no spam) nezumi.demon.co.uk> wrote in message
news:yXG2m.295$3o6.107 at (no spam) newsfe24.iad...
mlt wrote:
On this page its said that the gradient points in the direction of most
increase in intensity:
http://en.wikipedia.org/wiki/Image_gradient
but why are all the arrows pointing toward the black area? Is white not
defined as having the highest intensity?
And so the downhill direction is...
Regards,
Martin Brown
From the page:
"At each image point, the gradient vector points in the direction of largest
possible intensity increase, and the length of the gradient vector
corresponds to the rate of change in that direction"
If white is max an black is 0 then the arrow should point toward the white
area right? Unless start is where the arrowhead is located, but that would
not make sense.
[/quote:be21356240]
I'd tend to ignore that Wikipedia entry.
If you look at a maths dictionary, you will find at least two
definitions of 'gradient': (i) gradient in a direction; (ii) a gradient
vector.
For (i) think of climbing a hill, at any particular point you can travel
in an infinity of directions (in any angle 0--360 degrees). Each of
these directions has a gradient --- small distance ascended / small
distance travelled in horizontal direction (f2 -f1)/ (h2 - h1).
For (ii), think of the hill sitting a 2-d. coordinate system (x, y), x
and y defining a Cartesian plane.
Now we can have gradients in x- and y-directions --- think following
North (y) or West (x) in your hill-climbing.
The 'gradient vector' (defined according to unit vectors in the x- and
y-directions) comes from vectorially adding the two, i.e. the gradient
vector is
(Df/Dx, Df/Dy),
where D is a partial differential, f is the hill height (or image
intensity), i.e. DF/Dx = (f2 -f1)/ (x2 - x1) for a small movement from
x1 to x2.
The magnitude of that vector gives the maximum (sense (i) ) gradient at
the point.
The arrow is showing the gradient vector --- direction in the direction
of the arrow, and, usually, the length of the arrow may indicate magnitude.
Normally, in image processing as in hill climbing, I'd have the gradient
vector indicating increasing f (height), and that is the opposite to
what the Wikipedia article seems to indicate. Normally, white indicates
high intensity and black low ...
Gradient operators such as Roberts, Sobel, give discrete versions of
Df/Dx, Df/Dy).
Best regards,
Jon C.
--
Jonathan Campbell www.jgcampbell.com BT48, UK. |
|
|
| Back to top |
|
|
|
| Martin Leese... |
| Posted: Wed Jul 01, 2009 8:00 am |
|
|
|
Guest
|
mlt wrote:
[quote:fee973f5b2]On this page its said that the gradient points in the direction of most
increase in intensity:
http://en.wikipedia.org/wiki/Image_gradient
but why are all the arrows pointing toward the black area? Is white not
defined as having the highest intensity?
[/quote:fee973f5b2]
Read the article. It states that image
gradient is "an even gradation from low to
high values, as used from white to black in
the images to the right."
If it still bothers you then just copy the
Wikipedia images and take their negative.
--
Regards,
Martin Leese
E-mail: please at (no spam) see.Web.for.e-mail.INVALID
Web: http://members.tripod.com/martin_leese/ |
|
|
| Back to top |
|
|
|
| Skeptic... |
| Posted: Sat Jul 04, 2009 8:00 am |
|
|
|
Guest
|
On Jul 4, 6:12 am, iman islam <imanway2... at (no spam) gmail.com> wrote:
[quote:939d744299]What are the ‘Five Pillars’ of Islam?
[spam snipped][/quote:939d744299]
Strange. I didn't see spamming on the list. Yet here you are doing
it. Do you want us to consider all muslims as spammers? Is that what
your leaders want you to do? |
|
|
| Back to top |
|
|
|
|
|
All times are GMT - 5 Hours
The time now is Sun Nov 22, 2009 9:33 am
|
|