The MIFF header is composed entirely of ASCII characters. The fields in the header are keyword and value combination in the keyword=value format, with each keyword and value separated by an equal sign (=). Each keyword=value combination is delimited by at least one control or whitespace character. Comments may appear in the header section and are always delimited by braces. The MIFF header always ends with a colon (:) character, followed by a newline character. It is also common for a formfeed and a newline character to appear before the colon. You can then list the image keywords with more(1), without printing the binary image that follows the colon separator.
The following is a list of keyword=value combinations that may be found in a MIFF file:
Use this keyword when the image is a composite of a number of different tiles. A tile consists of an image and optionally a border and a label. <width> is the size in pixels of each individual tile in the horizontal direction and <height> is the size in the vertical direction. Each tile must have an equal number of pixels in width and equal in height. However, the width can differ from the height. <x offset> is the offset in number of pixels from the vertical edge of the composite image where the first tile of a row begins and <y offset> is the offset from the horizontal edge where the first tile of a column begins.
If this keyword is specified, a directory of tile names must follow the image header. The format of the directory is explained below.
The following is a sample MIFF header. In this example, <FF> is a formfeed character:
id=ImageMagick class=PseudoClass colors=256 signature=d79e1c308aa5bbcdeea8ed63df412da9 compression=RunlengthEncoded packets=27601 columns=1280 rows=1024 scene=1 { Rendered via Dore by Sandi Tennyson. } <FF> :
Note that keyword=value combinations may be separated by newlines or spaces and may occur in any order within the header. Comments (within braces) may appear anywhere before the colon.
If you specify the montage keyword in the header, follow the header with a directory of image tiles. This directory consists of a name for each tile of the composite image separated by a newline character. The list is terminated with a NULL character.
Following the header (or image directory if the montage keyword is in the header) is the binary image data itself. How the image data is formatted depends upon the class of the image as specified (or not specified) by the value of the class keyword in the header.
DirectClass images (class=DirectClass) are continuous-tone, RGB images stored as intensity values in red-green-blue order. Each color value is one byte in size for depth or 8 and there are three bytes per pixel (four with an optional matte value). If the depth is 16, each color value is two bytes with the most significant byte being first. The total number of pixels in a DirectClass image is calculates by multiplying the rows value by the column value in the header.
PseudoClass images (class=PseudoClass) are colormapped RGB images. The colormap is stored as a series of red-green-blue pixel values, each value being a byte in size. If the image depth is 16, each colormap entry is two bytes with the most significant byte being first. The number of colormap entries is indicated by the colors keyword in the header, with a maximum of 65,535 total entries allowed. The colormap data occurs immediately following the header (or image directory if the montage keyword is in the header).
PseudoClass image data is an array of index values into the color map. If there are 256 or fewer colors in the image, each byte of image data contains an index value. If the image contains more than 256 colors or the depth is 16, the index value is stored as two contiguous bytes with the most significant byte being first. The total number of pixels in a PseudoClass image is calculated by multiplying the rows value by the columns value in the header.
MIFF is capable of storing a digital signature for colormapped images. This signature was developed for use when animating a sequence of images on a colormapped X server. All of the signatures in a sequence of MIFF files are checked, and if they all match, you do not need to compute a global colormap.
The default colormap identifier is a digital signature computed using RSA's Data Security MD5 Digest Algorithm. (See a description of this algorithm in Internet draft, [MD5], July 1992. The colormap signature is computed if the MIFF file is part of a scene (i.e. the scene value does not equal 0).
The image data in a MIFF file may be uncompressed or may be compressed using one of two algorithms. The compression keyword in the header indicates how the image data is compressed. The run-length encoding (RLE) algorithm may be used to encode image data into packets of compressed data. For DirectClass images, runs of identical pixels values (not BYTE values) are encoded into a series of four-byte packets (five bytes if a matte value is included). The first three bytes of the packet contain the red, green, and blue values of the pixel in the run. The fourth byte contains the number of pixels in the run. This value is in the range of 0 to 255 and is one less than the actual number of pixels in the run. For example, a value of 127 indicates that there are 128 pixels in the run.
For PseudoClass images, the same RLE algorithm is used. Runs of identical index values are encoded into packets. Each packet contains the colormap index value followed by the number of index values in the run. The number of bytes n a PseudoClass RLE packet will be either two or three, depending upon the size of the index values. The number of RLE packets stored in the file is specified by the packets keyword in the header, but is not required.
A more complex algorithm, the Zip compression scheme, may be used to to achieve a greater compression ratio than run-length encoding. The number of compressed packets stored in the file is specified by the packets keyword in the header.
MIFF files may contain more than one image. Simply concatenate each individual image (composed of a header and image data) into one file.
Permission to use, copy, modify, distribute, and sell this software and its documentation for any purpose is hereby granted without fee, provided that the above copyright notice appear in all copies and that both that copyright notice and this permission notice appear in supporting documentation, and that the name of E. I. du Pont de Nemours and Company not be used in advertising or publicity pertaining to distribution of the software without specific, written prior permission. E. I. du Pont de Nemours and Company makes no representations about the suitability of this software for any purpose. It is provided "as is" without express or implied warranty.
E. I. du Pont de Nemours and Company disclaims all warranties with regard TO this software, including all implied warranties of merchantability and fitness, in no event shall E. I. du Pont de Nemours and Company be liable for any special, indirect or consequential damages or any damages whatsoever resulting from loss of use, data or profits, whether in an action of contract, negligence or other tortuous action, arising out of or in connection with the use or performance of this software.