functions in pnm.i - p

             #include "pnm.i"  
 
     makes pnm_read and pnm_write functions available, which see.  
keyword,  defined at i/pnm.i   line 28  

             color_image= pnm_colorize(image)  
 
     colorize the width-by-height array IMAGE, returning a 3-by-  
     width-by-height array, suitable for use with pnm_write.  
     The current graphics palette is used to perform the colorization.  
     The keywords cmin=, cmax=, and top= are recognized and have the  
     same meaning as for the plf, pli, and bytscl functions.  
interpreted function, defined at i/pnm.i   line 416  

             pnm_display, image  
 
     Attempt to display the IMAGE with the pli command and a pseudo-color  
     palette.  The IMAGE must be a 3-by-width-by-height array of RGB  
     pixel values.  If called as a function, the return value is the  
     width-by-height array of pixels which index into the new palette.  
     The palette is returned as the external variables red, green, and  
     blue.  The new palette is both coarse and slow to compute.  
     The size= keyword can be used to set the palette size.  The default  
     is 200 colors, the size of all the distribution palettes.  
     The flip=1 keyword can be used to flip the image bottom for top.  
     The square=1 limits flag is set; use  
        limits,square=0  
     to return to the default non-square plot limits.  
interpreted function, defined at i/pnm.i   line 219  

             image= pnm_read(filename)  
          or image= pnm_read(filename, noflip)  
 
     read a PBM, PGM, or PPM image from FILENAME.  PBM and PGM files  
     result in width-by-height arrays; PPM files result in 3-by-width-  
     by-height arrays.  If NOFLIP is present and non-zero, the returned  
     image is in exactly the order stored in FILENAME.  Otherwise, the  
     height dimension is reversed, since images in the files read top  
     to bottom, while the Yorick image plotting commands go bottom to  
     top.  
interpreted function, defined at i/pnm.i   line 35  

             pnm_write, image, filename)  
          or pnm_write, image, filename, noflip)  
 
     write IMAGE to a PBM, PGM, or PPM file called FILENAME.  If  
     NOFLIP is present and non-zero, the image in FILENAME is stored  
     in exactly the order of IMAGE in memory.  Otherwise, the height  
     dimension is reversed, since images in the files read top to  
     bottom, while the Yorick image plotting commands go bottom to  
     top.  
     If IMAGE is a 3-by-width-by-height array, a PPM will be written.  
     Otherwise, IMAGE must be a width-by-height array.  If IMAGE  
     contains at most two distinct values, a PBM will be written,  
     otherwise a PGM.  You can force a PGM using the bits=8 keyword.  
     If IMAGE is of type char, it will be used as is, otherwise it  
     will be scaled to the range 0 to 255.  (Note that for a PPM, this  
     means the largest single rgb component value sets the scale.)  
     If the text= keyword is present and non-zero, a text PNM file  
     will be written; the default is to write a binary or raw PNM.  
     If the noscale=1 keyword is supplied, the IMAGE will not be  
     scaled to the range 0 to 255.  In this case, the IMAGE must  
     have an integer data type with minimum value >=0.  Note that  
     if max(image)>255, a text PGM or PPM file will be written  
     (overriding the text= keyword).  Raw PNM is char-only.  
     If bits= is not specified, then noscale=1 forces pnm_write to  
     guess the value you intended the brightest component value.  
     If IMAGE is not color, the guess is bits=1 if there are only  
     1 or 2 values, bits=8 if max(IMAGE)<256, and bits=16 otherwise.  
     For color IMAGE, the guess is either bits=8 or bits=16.  
     Normally, PGM and PPM files have a pixel or color component  
     size of 8 bits, so they run from 0 (darkest) to 255 (brightest).  
     With bits= you can force a different maximum value.  Use bits=8  
     to force a binary IMAGE to be written as a PGM.  The largest  
     legal value of bits is 16, the smallest is 1.  
interpreted function, defined at i/pnm.i   line 287