Legacy ImageMagick Discussions Archive

I have a similar problem, if I try to write tiff files after performin a rotation with the original file. It happens if the original file is a png file or a tiff file.

rose@moose:/home/rose(405)$ convert -rotate 10 OrigImages/lena.png OutImages/lena.tiff
convert: tif_dirwrite.c:2084 ...

BTW, can anybody give me a hint, which algorithms is used, that some pixel values become negative or larger than the former maximal value (255)?

If I rotate an image I get values for image depth, minimum and maximum, which I does not expect. I use the following rotate-image_with_getChannelRange.c program:


#include <stdio.h>
#include <stdlib.h>
#include <wand/MagickWand.h>

print_stat(MagickWand *wand, long long unsigned max_img_val, long ...

The result is now:

rose@moose:/home_moose/rose/Txt/src/Test/C/ImageMagick/Wand(268)$ ./getChannelRange lena.png
IM_quantum_depth_str=|Q32| IM_quantum_depth=|32| img_depth=8 quantum_format=|(null)|
RedChannel=1
GreenChannel=2
BlueChannel=4
after 'GetImageChannelRange(,RedChannel,) ' minimum= 38.0 ...

I assume, that I found the solution:


#include <stdio.h>
#include <stdlib.h>
#include <wand/MagickWand.h>

int main(int argc,char **argv) {

#define ThrowWandException(wand) { \
description=MagickGetException(wand,&severity);\
(void) fprintf(stderr,"%s %s %lu %s\n",GetMagickModule(),description ...

With 'identify -verbose file' I can show minmal and maximal pixel values, e.g.

rose@moose:/home_moose/rose/Txt/src/Test/C/ImageMagick/Wand(203)$ identify -verbose lena.png
...
Colorspace: RGB
Depth: 8-bit
Channel depth:
red: 8-bit
green: 8-bit
blue: 8-bit
Channel statistics:
Red:
min: 38 ...

I repeated the test with ImageMagick 6.7.4-0:

rose@moose:/home_moose/rose/Txt/projects/MicroElectrodes/Test/GetOneAuthenticPixel_versus_MagickExportImagePixels(741)$ identify -version
Version: ImageMagick 6.7.4-0 2011-12-12 Q32 http://www.imagemagick.org
Copyright: Copyright (C) 1999-2011 ...

But what about the main problem the identification of quantum:format?

It seems that
const char *option=GetImageOption(image_info,"quantum:format");
does not work for me. I use the following source:


#include <stdio.h>
#include <string.h>
#include <magick/MagickCore.h>

int main (int argc, char *argv[]) {
MagickCoreGenesis(*argv,MagickTrue);
ExceptionInfo ...

Thanks magick,

I will wait for GetQuantumFormat(). But in the meantime is there any other solution to get information from the Core or Wand API whether a image has long or float pixel?

Regards Juergeen

I have some pictures, where 'identify -verbose' says "quantum:format: floating-point". I want to save the pixels of these and other images with GetOneAuthenticPixel in MagickCore and with MagickExportImagePixels in MagickWand to a long or a double array. To do this, I want to decide from the ...