Search found 88 matches
- 2012-01-07T09:56:58-07:00
- Forum: Digital Image Processing
- Topic: proper scaling of the Jinc filter for EWA use
- Replies: 234
- Views: 618000
Re: proper scaling of the Jinc filter for EWA use
This is not lens correction, it's motion blur and focussing error. ( If you're interested in that, look at GIMP's lens "distortion" , which can be used with negative values as a lens corrections. IIRC, it's a matrix op rather than convo ). very briefly, motion blur is a convolution with a ...
- 2012-01-07T06:36:54-07:00
- Forum: Digital Image Processing
- Topic: proper scaling of the Jinc filter for EWA use
- Replies: 234
- Views: 618000
Re: proper scaling of the Jinc filter for EWA use
lanczos3=" lz=zz/4.0; (abs(lz)<1) ? sinc(zz) / sinc(lz) : white" Just a quick aside. After much head banging I finally realised that to do the same thing in a deblur context, where the jinc or sinc is in the denominator , you need to divide by the window function not multiply. Seems embar...
- 2012-01-04T20:35:04-07:00
- Forum: Bugs
- Topic: bad error trapping could cause unexpected shell commands
- Replies: 8
- Views: 17201
Re: bad error trapping could cause unexpected shell commands
$ convert -size $((5*$wid))x$((5*$ht)) xc: -tile (-extract 50%x25%+25%+5% tile.png)\ bash: syntax error near unexpected token `(' $ /tmpd/img $ -draw "rectangle 0,0 $((5*$wid)),$((5*$ht))" \ > output-tile.png There was no spaces on what I posted but you are in essence correct . The shell ...
- 2012-01-04T16:13:15-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
Yes magnitude and phase is a polar co-ord system, the complex plane in a cartesian representation. The two are equivalent expressions of the same thing, so a result in one frame cannot confirm a result in the other , it's the same thing. I still believe the fact that I can reproduce the mag/phase im...
- 2012-01-04T13:35:20-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
But I really don't know how to interpret the phase profile. It seems to have very high frequency oscillations within slower osciallations, the later of which seem to match the cycles in the mag profile. I suspect the hf is spurious, I think the correct phase is a linear ramp. I know that FFTW is a ...
- 2012-01-04T13:25:16-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
Oh please, don't quote WP as a reference for anything. It's like watching a Hollywood film "based on a true story" and imagining you are learning history. So are you saying that it is wrong. A lot of experts have contributed to Wikipedia. So there must be something to it. I cannot explain...
- 2012-01-04T03:39:53-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
http://www.fmwconcepts.com/misc_tests/FFT_v6744/square31_hdri_fft_mp-1.png now If I take a x=0 or y=0 section through that plot , it's about what I would expect: an even cyclic variation of phase. The light / dark discontinuity is in fact the -pi , +pi wrap around. I was expecting something more li...
- 2012-01-04T02:55:29-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
quote and link in first post. According to what I see and what you confirmed, it seems HDRI and non_HDRI are now scaled the same and it's not 0-2*pi. That is correct. They are both scaled to 0 to quantumrange so that the phase if easily visible without scaling. Internally they are used properly wit...
- 2012-01-04T02:47:27-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
The other issue I am still puzzled about is how these real and imaginary components relate to the fact that in the 1D (analog) fourier transform, the rect and sinc are transform pairs. No one ever states anything about the imaginary part. See table in http://en.wikipedia.org/wiki/Fourier_transforms...
- 2012-01-03T18:19:40-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
quote and link in first post. According to what I see and what you confirmed, it seems HDRI and non_HDRI are now scaled the same and it's not 0-2*pi.
- 2012-01-03T17:47:22-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
What part of the doc is broken -- where specifically? specifically the bit I noted above about non-HDRI being different. In fact non-HDRI seems to be as stated but the implication that HDRI is different has apparently not been the case " for a long time". When I went to the doc the try to...
- 2012-01-03T14:39:42-07:00
- Forum: Bugs
- Topic: crop and extact to png
- Replies: 4
- Views: 9818
Re: crop and extact to png
It does indeed.
IM has a learning curve like the foothills of the Himalayas , looks like I'm getting out of breath !
Many thanks.
IM has a learning curve like the foothills of the Himalayas , looks like I'm getting out of breath !
Many thanks.
- 2012-01-03T14:31:39-07:00
- Forum: Bugs
- Topic: FFT phase ?
- Replies: 19
- Views: 39672
Re: FFT phase ?
No roll is necessary for the image, unless you are using it as the filter function. Then it has to be rolled so as centered at 0,0. In my scripts such as fftconvol fftdeconvol, I do the roll of the spatial filter (only), not the image, internal to the script. So it is not necessary to do the roll e...
- 2012-01-03T13:20:19-07:00
- Forum: Bugs
- Topic: crop and extact to png
- Replies: 4
- Views: 9818
Re: crop and extact to png
many thanks for clearing that up.
How should I have know to do this? I have gone over the doc for -extract and crop many times now and see no indication that there is some residual information nor of the need to "repage".
How should I have know to do this? I have gone over the doc for -extract and crop many times now and see no indication that there is some residual information nor of the need to "repage".
- 2012-01-03T11:08:20-07:00
- Forum: Bugs
- Topic: crop and extact to png
- Replies: 4
- Views: 9818
crop and extact to png
The above was created by the following command:
Code: Select all
convert -gravity NorthEast -crop "600x600+0+0" biche.jpg biche.png
IM bug, GIMP bug or png spec abiguity?
thx