From 1fd361a1ea06e44286c213ca1f814f49306fdc43 Mon Sep 17 00:00:00 2001 From: giraffedata Date: Sat, 19 Aug 2006 03:12:28 +0000 Subject: Create Subversion repository git-svn-id: http://svn.code.sf.net/p/netpbm/code/trunk@1 9d0c8265-081b-0410-96cb-a4ca84ce46f8 --- converter/ppm/ppmtoyuvsplit.c | 197 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 197 insertions(+) create mode 100644 converter/ppm/ppmtoyuvsplit.c (limited to 'converter/ppm/ppmtoyuvsplit.c') diff --git a/converter/ppm/ppmtoyuvsplit.c b/converter/ppm/ppmtoyuvsplit.c new file mode 100644 index 00000000..2dddebfc --- /dev/null +++ b/converter/ppm/ppmtoyuvsplit.c @@ -0,0 +1,197 @@ +/* ppmtoyuvsplit.c - convert a portable pixmap into 3 raw files: +** - basename.Y : The Luminance chunk at the size of the Image +** - basename.U : The Chrominance chunk U at 1/4 +** - basename.V : The Chrominance chunk V at 1/4 +** The subsampled U and V values are made by arithmetic mean. +** +** If CCIR601 is defined, the produced YUV triples are scaled again +** to fit into the smaller range of values for this standard. +** +** by A.Beck +** Internet: Andre_Beck@IRS.Inf.TU-Dresden.de +** +** Based on ppmtoyuv.c +** +** Permission to use, copy, modify, and distribute this software and its +** documentation for any purpose and without fee is hereby granted, provided +** that the above copyright notice appear in all copies and that both that +** copyright notice and this permission notice appear in supporting +** documentation. This software is provided "as is" without express or +** implied warranty. +*/ + +/* Wether to create YUV in JFIF(JPEG) or CCIR.601(MPEG) scale */ +#define CCIR601 + +/* Wether to use pm_close() or fake it -- don't ask me why */ +/* #define ORIGINAL */ + +/* ALPHA Kludge by Franky */ + +#ifdef __alpha +#define myLONG int +#else +#define myLONG long +#endif + +#include +#include "ppm.h" + +int +main(argc, argv) +char **argv; +{ + FILE *ifp,*vf,*uf,*yf; + pixel *pixelrow1,*pixelrow2; + register pixel *pP1,*pP2; + int rows, cols, format, row; + register int col; + pixval maxval; + myLONG u,v,y0,y1,y2,y3,u0,u1,u2,u3,v0,v1,v2,v3; + unsigned char *y1buf,*y2buf,*ubuf,*vbuf; + char ufname[256],vfname[256],yfname[256]; + + + ppm_init(&argc, argv); + + if ((argc>3)||(argc<2)) pm_usage("basename [ppmfile]"); + + if (argc == 3) ifp = pm_openr(argv[2]); + else ifp = stdin; + + strcpy(ufname,argv[1]); + strcpy(vfname,argv[1]); + strcpy(yfname,argv[1]); + + strcat(ufname,".U"); + strcat(vfname,".V"); + strcat(yfname,".Y"); + + uf = fopen(ufname,"wb"); + vf = fopen(vfname,"wb"); + yf = fopen(yfname,"wb"); + + if(!(uf && vf && yf)) { + perror("error opening output files"); + exit(0); + } + ppm_readppminit(ifp, &cols, &rows, &maxval, &format); + + if(cols & 1) fprintf(stderr, + "%s: Warning: odd columns count, exceed ignored\n", + argv[0]); + if(rows & 1) fprintf(stderr, + "%s: Warning: odd rows count, exceed ignored\n", + argv[0]); + + pixelrow1 = ((pixel*) pm_allocrow( cols, sizeof(pixel) )); + pixelrow2 = ((pixel*) pm_allocrow( cols, sizeof(pixel) )); + + y1buf = (unsigned char *) pm_allocrow( cols, 1 ); + y2buf = (unsigned char *) pm_allocrow( cols, 1 ); + ubuf = (unsigned char *) pm_allocrow( cols, 1 ); + vbuf = (unsigned char *) pm_allocrow( cols, 1 ); + + for (row = 0; row < (rows & ~1); row += 2) { + unsigned char *y1ptr,*y2ptr,*uptr,*vptr; + + ppm_readppmrow(ifp, pixelrow1, cols, maxval, format); + ppm_readppmrow(ifp, pixelrow2, cols, maxval, format); + + pP1 = pixelrow1; pP2 = pixelrow2; + y1ptr = y1buf; y2ptr = y2buf; vptr = vbuf; uptr = ubuf; + + for (col = 0 ; col < (cols & ~1); col += 2) { + pixval r0,g0,b0,r1,g1,b1,r2,g2,b2,r3,g3,b3; + + /* first pixel */ + r0 = PPM_GETR(*pP1); + g0 = PPM_GETG(*pP1); + b0 = PPM_GETB(*pP1); + pP1++; + /* 2nd pixel */ + r1 = PPM_GETR(*pP1); + g1 = PPM_GETG(*pP1); + b1 = PPM_GETB(*pP1); + pP1++; + /* 3rd pixel */ + r2 = PPM_GETR(*pP2); + g2 = PPM_GETG(*pP2); + b2 = PPM_GETB(*pP2); + pP2++; + /* 4th pixel */ + r3 = PPM_GETR(*pP2); + g3 = PPM_GETG(*pP2); + b3 = PPM_GETB(*pP2); + pP2++; + + +/* The JFIF RGB to YUV Matrix for $00010000 = 1.0 + +[Y] [19595 38469 7471][R] +[U] = [-11056 -21712 32768][G] +[V] [32768 -27440 -5328][B] + +*/ + + y0 = 19595 * r0 + 38469 * g0 + 7471 * b0; + u0 = -11056 * r0 - 21712 * g0 + 32768 * b0; + v0 = 32768 * r0 - 27440 * g0 - 5328 * b0; + + y1 = 19595 * r1 + 38469 * g1 + 7471 * b1; + u1 = -11056 * r1 - 21712 * g1 + 32768 * b1; + v1 = 32768 * r1 - 27440 * g1 - 5328 * b1; + + y2 = 19595 * r2 + 38469 * g2 + 7471 * b2; + u2 = -11056 * r2 - 21712 * g2 + 32768 * b2; + v2 = 32768 * r2 - 27440 * g2 - 5328 * b2; + + y3 = 19595 * r3 + 38469 * g3 + 7471 * b3; + u3 = -11056 * r3 - 21712 * g3 + 32768 * b3; + v3 = 32768 * r3 - 27440 * g3 - 5328 * b3; + + /* mean the chroma for subsampling */ + + u = (u0+u1+u2+u3)>>2; + v = (v0+v1+v2+v3)>>2; + +#ifdef CCIR601 + + y0 = (y0 * 219)/255 + 1048576; + y1 = (y1 * 219)/255 + 1048576; + y2 = (y2 * 219)/255 + 1048576; + y3 = (y3 * 219)/255 + 1048576; + + u = (u * 224)/255 ; + v = (v * 224)/255 ; +#endif + + + *y1ptr++ = (y0 >> 16) ; + *y1ptr++ = (y1 >> 16) ; + *y2ptr++ = (y2 >> 16) ; + *y2ptr++ = (y3 >> 16) ; + + + *uptr++ = (u >> 16)+128 ; + *vptr++ = (v >> 16)+128 ; + + } + fwrite(y1buf, (cols & ~1), 1, yf); + fwrite(y2buf, (cols & ~1), 1, yf); + fwrite(ubuf, cols/2, 1, uf); + fwrite(vbuf, cols/2, 1, vf); + } + +/* I dunno why pm_close sees an error... get rid of it */ + +#ifdef ORIGINAL + pm_close(ifp); +#else + if(ifp != stdin) fclose(ifp); +#endif + fclose(yf); + fclose(uf); + fclose(vf); + exit(0); +} -- cgit 1.4.1