Re: image normalisation




On Feb 11, 1:25 am, "ImageAnalyst" <imageanal...@xxxxxxxxxxxxxx>
wrote:
Don't know if it will help with your images but check into homomorphic
filtering (do a web search). Or maybe simply just apply a LUT (lookup
table) to your image to remap the intensities to something less
bright. Or there are various histogram matching algorithms if you
need to match your image's histogram to some other image's
intensities.

i'll check it out.


Are your images basically snapshots where the sun could be at any
angle casting shadows in weird directions and along irregularly shaped
objects? Like the sun shining down on the Manhattan cityscape. Or
does the brightness always come from some fixed, predefined angle? I
doubt your images are such that you could simply find the bright
areas, create a mask from them, then shift the mask and make the areas
in the shifted mask darker - that would either be incredibly lucky or
your have some kind of controlled image acquisition situation.

the case is that i should be possible to make zhe algorithm work on
several picture databases. my first problems showed when i used such
one. the pictures were taken with the light from down-right, and it
was impossible to extract anything. on another one the pictures were
too dark. at the first step it was also impossible, but changing the
threshold value in the algorithm it worked. The algorithm is great.
but i have to find a way normalise the pictures so i can use the
"same" value.
something like if i have one picture as a template, normalise the
other pictures according to the template.



I have no idea what you mean by the pixels are "clusterized" - I've
never heard that term before so please explain.
the pixels on the picture look like mosaic stones. there is a too big
value difference between pixels in the neighbourhood. so by using a
normal gaussian edge detect, i should get back only the boder but i
get inside the wished edges all pixels indetified like border but it
looks more like noise. just pixelized.



Is that camera so bad
that the sensor is not in a standard periodic array like Bayer Patter
or honeycomb (Fujifilm)? That would be highly unusual. Of course if
you have access to the camera, it's better to just replace it and get
good images to start with rather than go to algorithmic heroics to try
to fix crappy images. But sometimes you are stuck with poor images
that others shot (vacation snapshots, etc.) and it's your job to do
your best with what you got, bad as they may be.

Thats happens when idiots take a camera and give you images. I come to
imageprocessing after a couple of years on gimp because i want more
freedom then using someguys plugins. i can solve the quality in gimp,
but i'm not allowed to do it. Do you know some place where i can
download a good quality face picture database.

I don't want to give up this algorithm. it works fantastic (my work).
there are guys became doctors with some crappy algorithm for eye-
extraction, some on lips, some eyebrows, but i can extract on a
"controled" image, the eyes, mouth, eyebrows, but only the nose is
showing some problems.

Good luck,
ImageAnalyst


On Feb 10, 4:29 pm, "azrael" <jura.gro...@xxxxxxxxx> wrote:

hy guys. im working on a aplicaton for image features extraction. In
the beginning i was convinced that tghis would not be hard, but as
soon i got the pictures that i should be working on I was shocked.
they are taken with a bad camera so the pixels are too clusterized and
there is too much noise. it is impossible to get any information from
the image when i apply any edge detect or corner detect algorithm. I
tried bluring and then extracting the edges. nothing. so i used some
combinations of algorithms and I found an algorithm combination that
works well but there is (as usual) problem. it's too sensitive to
brightness. Is there anything you would recomend. i need a
normalisation for brightness. not only to bring the image to one, but
is there anything to correct the brightness from a specific direction,
or.


.

Relevant Pages

  • Re: image normalisation
    ... Sounds like you need to look into histogram matching techniques (good ... does the brightness always come from some fixed, ... the pictures were taken with the light from down-right, ... threshold value in the algorithm it worked. ...
    (sci.image.processing)
  • Re: image normalisation
    ... tried the pictures they tried on. ... to admit, that his algorithm may might work, but only if he would ... does the brightness always come from some fixed, ... something like if i have one picture as a template, normalise the ...
    (sci.image.processing)
  • Re: image normalisation
    ... i saw a dr. work from some guy about eye extraction. ... tried the pictures they tried on. ... to admit, that his algorithm may might work, but only if he would ... does the brightness always come from some fixed, ...
    (sci.image.processing)
  • Re: image normalisation
    ... the beginning i was convinced that tghis would not be hard, ... soon i got the pictures that i should be working on I was shocked. ... the image when i apply any edge detect or corner detect algorithm. ... normalisation for brightness. ...
    (sci.image.processing)
  • Changing Pictures Contrast
    ... I have two pictures in a do and I want to change the contrast/ ... brightness of each programatically. ... find the two images via vba and second how can I set the brightness & ...
    (microsoft.public.word.vba.general)
Loading