Re: need some help with custom filter in Photoshop

From: Blaine Waddington (v_at_z.m)
Date: 02/27/05 

Date: Sun, 27 Feb 2005 22:51:28 GMT

Hi Guys:

Boy, this seems to be some tough math here but I think I'm beginning to
understand it. Let me try to phrase what I think is happening:

If I have a 7x7 kernal, for example, and apply that to a 500x500 pixel
image, each pixel will be changed according to the configuration of the 7x7,
or would this be each 7x7 pixel area being changed?

I inputted the values for vertical lines and such, as well as tried selected
blurs in only certain directions. I think I'm beginning to understand the
process. Let me try to understand a bit more with the following question:

Suppose I have a tiny 5x5 pixel image for 25 square pixels. I apply a 7x7
matrix. Is this matrix too large for this tiny image? Should my image have
been at least a 7x7 size as well?

Thanks,
Blaine

"Anders Thulin" <ath_no_spam_please@algonet.se> wrote in message
news:jKfUd.18369$d5.142993@newsb.telia.net...
> Blaine Waddington wrote:
>
> > I keep running into these "custom" or kernal filters in Photoshop and
other
> > programs but, try as I might, I have been unable to find good tutorials
as
> > to how they work with some practical input values.
>
> A textbook in image processing perhaps? *Real* image processing, not
> image editing or synthesizing. (I remember Gonzales & Wintz from way back,
but
> Gonzales & Woods probably replaces that nowadays.)
>
> Kernels are square arrays (typically of odd sizes, like 3x3) containing
a
> number of specially selected coeffiecents. The size is the 'input area',
> the coefficients the weighting of the corresponding pixels in the input
image,
> and the actual function being performed over these pixels is typically
> summing follwoed by a scaling operation implicit in the kernel. Output is
> one single pixel orresponding to the center of the kernel.
>
> Processing is done by placing the kernel in all possible positions on
the
> input image, and creating an output image from the input pixels. (image
> edges get special treatment.) For example (3x3 kernel, with coordinate
[0,0] in the center):
>
> out[x,y] := sum of (in[x-1,y-1]*kernel[-1,-1], ...,
in[x+1,y+1]*kernel[1,1]);
>
> Each pixel gets multiplied with the corresponding kernel coefficient, and
the result is
> summed.
>
> If the kernel is all 1, and you add a division by 9 at the end, this is
a simple
> averaging filter (remember, pixels are from input image only). By changing
the kernel,
> you can weight pixels to influence the outcome more (though you may need
to change the
> 9 accordingly). For instance, corner pixels may get weight 1, orthogonal
pixels weight
> 4, central pixel weight 9 -- this filter is slightly more spatially
sensitive than
> the 'all 1' one.
>
> You can also use kernels for feature detection, such as edge detection.
>
> If kernel is [[-1,-1,-1],[-1,8,-1],[-1,-1,-1]], you'll get a
> resulting image that is all edges from the original image, i.e. places
> where a low pixel value was next to a high pixel, regardless of direction
> (though within the spatial limitations of a 3x3 kernel).
>
> If you change that to [-1,8,-1][-1,8,-1][-1,8,-1], you'll
> enhance narrow vertical lines. Typically, the results from several
> such 'convolutions' are combined for further processing, and possibly
> even used as input images for higher-level convolutions.
>
> 3x3 pixels are a bit too primitive, though -- 5x5 or 7x7 often makes
for
> better results, but of course requires much more processing power.
> For these sizes, you need hardware assist ... or specialized hardware.
> In a past life, I worked with a company where 11x11 kernels were thought
> of as barely useful -- 19x19 and above were more like real life.
>
> If you want to go deeper, look for the term 'convolution' in the
context
> of image processing. You may need some math background, though.
>
>
> --
> Anders Thulin ath*algonet.se http://www.algonet.se/~ath

Relevant Pages

  • Magic kernel for image zoom resampling?
    ... My need was to downsample an image by a factor of 8 in each direction ... To downsample or upsample by a factor of 2, use a kernel ... placed over every second pixel in each direction. ... The results for upsampling are astoundingly good, ...
    (sci.image.processing)
  • Re: Selective Gaussian -> Early test app posted
    ... >> Well, the problem with an arbitrary kernel, is that it imposes a ... >> gaussian kernel is separable, you can take advantage of this fact in ... can we take advantage of this for selective gaussian blur? ... is the value of some color component of a pixel at ...
    (borland.public.delphi.language.basm)
  • Re: Rookie having problems with some filter code. Any help?
    ... The negative coefficients in your kernel ... The kernel you have here is an edge detection filter. ... input pixel value. ... the signed int filter result to the source pixel and then apply clipping. ...
    (sci.image.processing)
  • Re: Magic kernel for image zoom resampling?
    ... I am not interlacing this kernel with identity for enlargement. ... Start with downsampling. ... pixel outside the image; that's an edge effect that we can deal with. ... Gaussian and this has been covered in literature, ...
    (sci.image.processing)
  • Re: Magic kernel for image zoom resampling?
    ... I am not interlacing this kernel with identity for enlargement. ... Start with downsampling. ... pixel outside the image; that's an edge effect that we can deal with. ... For upsampling, it's simplest to imagine that we're reversing the above ...
    (comp.graphics.algorithms)