going from Inverse discrete FFT to spectrogram - c++

Question

I need to locate some generic C++ library that takes the inverse fft output (fftw_complex format, i.e. two doubles) and converts this data to an image file such as png. I can waterfall the dffts to obtain the 2d data (and use 10log10(rere+imim) to obtain magnitudes for each frequency component) but I don't know which image library will work.
I did use an older program called zimage at one time, but it seems no longer available. I do not have MATLAB on my Ubuntu 9.10 system (but I do have Octave)
Can Octave generate the waterfall images? I also need to convert the spectrogram into a wav sound file too.
Any ideas??

Answer 1

The easiest image format to create is PNM. You can print it as a text file, and then convert it using most graphics programs. Here is an example from the Wikipedia page:
P2 24 7 15
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 3 3 3 3 0 0 7 7 7 7 0 0 11 11 11 11 0 0 15 15 15 15 0
0 3 0 0 0 0 0 7 0 0 0 0 0 11 0 0 0 0 0 15 0 0 15 0
0 3 3 3 0 0 0 7 7 7 0 0 0 11 11 11 0 0 0 15 15 15 15 0
0 3 0 0 0 0 0 7 0 0 0 0 0 11 0 0 0 0 0 15 0 0 0 0
0 3 0 0 0 0 0 7 7 7 7 0 0 11 11 11 11 0 0 15 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Save that text in a file named "feep.pgm", and you'll see what I mean.
http://en.wikipedia.org/wiki/Netpbm_format
You'll have to scale your 10log10 information to pixel values.

Answer 2

OpenCV is one library that can handle PNG files, among a variety of other formats. It should be readily available on your Ubuntu 9.10 system using apt get libcv-dev (from memory, you may have to double-check the package name).
/*
* compile with:
*
* g++ -Wall -ggdb -I. -I/usr/include/opencv -L /usr/lib -lm -lcv -lhighgui -lcvaux filename.cpp -o filename.out
*/
#include <cv.h>
#include <highgui.h>
/*
* Your image dimensions.
*/
int width;
int height;
CvSize size = cvSize(width, height);
/*
* Create 3-channel image, unsigned 8-bit per channel.
*/
IplImage *image = cvCreateImage(size, IPL_DEPTH_8U, 3);
for (int i = 0; i < width; ++i)
for (int j = 0; j < height; ++j)
{
unsigned int r;
unsigned int g;
unsigned int b;
/*
* Keep in mind that OpenCV stores things in BGR order.
*/
CvScalar bgr = cvScalar(b, g, r);
cvSet2D(image, i, j, bgr);
}
cvSaveImage("filename.png", image);
cvReleaseImage(&image);