/*************************************************************************** * This file is a part of CADS/UVS fits2jpeg conversion software * * Copyright (C) 2012 by CADS/UV Software Team, * * Indian Institute of Astrophysics * * Bangalore 560034 * * cads_AT_iiap.res.in * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. * ***************************************************************************/ /*Header Definitions*/ #include "fits2jpeg.h" void scale_image(int scale, int npixels, float *data, JSAMPLE *image_buffer) { unsigned int i = 0; int JMAXVAL = 255; float datamax = 0.0, datamin = 0.0, tmp = 0.0; float hist[256] = {0.0}, cumhist[256] = {0.0}; float scl_data = 0.0; /* first find min & max in data */ datamax = -1.0 * FLT_MAX; datamin = FLT_MAX; for (i = 0; i < npixels; ++i) { if (data[i] > datamax) datamax = data[i]; if (data[i] < datamin) datamin = data[i]; } /*endfor*/ /* Convert data into bytscaled values for jpeg file */ /* the dynamic range is reduced to 255 for jpeg */ scl_data = (datamax - datamin)/(float)JMAXVAL; for (i = 0; i < npixels; ++i) data[i] = (data[i] - datamin)/scl_data; /* All data is now squeezed into the range 0 - 255 */ /* NOTE: At this point onwards min & max is 0 and 255 respectively */ datamax = (float)JMAXVAL; datamin = 0.0; /* initialize image histogram. ensure all are zeroes in hist[] */ /*-------------------------------------------------------------------*/ for (i = 0; i <= JMAXVAL; ++i) hist[i] = 0; /* construct the image histogram */ tmp = 1.0/(float)npixels; for (i = 0; i <= npixels; ++i) hist[(int)floor(data[i])] += tmp; /* And the cumulative histogram */ cumhist[0] = hist[0]; for (i = 1; i <= JMAXVAL; ++i) cumhist[i] += cumhist[i - 1] + hist[i]; /* Linear scale (min-max) : This is the default scaling * if we dont generate image_buffer here, histo-eq will fail */ for (i = 0; i < npixels; ++i) image_buffer[i] = (int)(data[i]); /*-------------------------------------------------------------------*/ switch (scale) { case 1 : /* Square root */ printinfo("Using square-root scale"); scl_data = sqrt((float)JMAXVAL)/(float)JMAXVAL; for (i = 0; i < npixels; ++i) image_buffer[i] = (int)(sqrt(data[i])/scl_data); break; case 2 : /* Square */ printinfo("Using quadratic scale"); scl_data = pow((float)JMAXVAL,2)/(float)JMAXVAL; for (i = 0; i < npixels; ++i) image_buffer[i] = (int)abs((pow(data[i],2) - 1.0)/scl_data); break; case 3 : /* Cubic */ printinfo("Using cubic scale"); scl_data = pow((float)JMAXVAL,3)/(float)JMAXVAL; for (i = 0; i < npixels; ++i) image_buffer[i] = (int)abs((pow(data[i],3) - 1.0)/scl_data); break; case 4 : /* log */ printinfo("Using log scale"); scl_data = log(1.0 + (float)JMAXVAL)/(float)JMAXVAL; for (i = 0; i < npixels; ++i) image_buffer[i] = (int)((log(abs(data[i]) + 1.0))/scl_data); break; case 5 : /* contrast stretch */ printinfo("Performing histogram stretch (normalization)"); /* We need to go through the cumulative histogram to pick the * appropriate values for datamin and datamax */ i = 0; while (i < JMAXVAL) { if (cumhist[i] >= 0.01) { datamin = (float) i; break; } i++; } i = JMAXVAL; while (i > 0) { if (cumhist[i] <= 0.99) { datamax = (float) i; break; } i--; } scl_data = (datamax - datamin)/(float)JMAXVAL; for (i = 0; i < npixels; ++i) { if (image_buffer[i] >= datamax) image_buffer[i] = JMAXVAL; else if (image_buffer[i] <= datamin) image_buffer[i] = 0; else image_buffer[i] = (int) abs((image_buffer[i] - datamin)/scl_data); } break; case 6 : /* histogram equalization */ printinfo("Performing Histogram Equalization"); for (i = 0; i < npixels; ++i) image_buffer[i] = cumhist[image_buffer[i]] * 255; break; default : printinfo("Using linear scale"); break; } }