1Tifftopnm User Manual(0) Tifftopnm User Manual(0)
2
6 tifftopnm - convert a TIFF file into a PNM image
7
10 tifftopnm
11 [-alphaout={alpha-filename,-}] [-headerdump] [-verbose] [-respectfil‐
13 lorder] [-byrow] [-orientraw] [tiff-filename]
14
18 This program is part of Netpbm(1).
19 tifftopnm reads a TIFF file as input and produces a PNM image as out‐
21 put. The type of the output file depends on the input file - if it's
22 black & white, generates a PBM image; if it's grayscale, generates a
23 PGM image; otherwise, a PPM image. The program tells you which type it
24 is writing.
25 If the TIFF file contains multiple images (multiple 'directories,'),
27 tifftopnm generates a multi-image PNM output stream. Before Netpbm
28 10.27 (March 2005), however, it would just ignore all but the first
29 input image.
30 The tiff-filename argument names the regular file that contains the
32 Tiff image. If you specify '-' or don't specify this argument,
33 tfftopnm uses Standard Input. In either case, the file must be seek‐
34 able. That means no pipe, but any regular file is fine.
35 TIFF Capability
38 pamtotiff uses the Libtiff.org TIFF library (or whatever equivalent you
39 provide) to interpret the TIFF input. So the set of files it is able
40 to interpret is determined mostly by that library.
41 This program cannot read every possible TIFF file -- there are myriad
43 variations of the TIFF format. However, it does understand monochrome
44 and gray scale, RGB, RGBA (red/green/blue with alpha channel), CMYK
45 (Cyan-Magenta-Yellow-Black ink color separation), and color palette
46 TIFF files. An RGB file can have either single plane (interleaved)
47 color or multiple plane format. The program reads 1-8 and 16 bit-per-
48 sample input, the latter in either bigendian or littlendian encoding.
49 Tiff directory information may also be either bigendian or littlendian.
50 There are many TIFF formats that tifftopnm can read only if the image
52 is small enough to fit in memory. tifftopnm uses the TIFF library's
53 TIFFRGBAImageGet() function to process the TIFF image if it can get
54 enough memory for TIFFRGBAImageGet() to store the whole image in memory
55 at once (that's what TIFFRGBAImageGet() does). If not, tifftopnm uses
56 a more primitive row-by-row conversion strategy using the raw data
57 returned by TIFFReadScanLine() and native intelligence. That native
58 intelligence does not know as many formats as TIFFRGBAImageGet() does.
59 And certain compressed formats simply cannot be read with TIFFReadScan‐
60 Line().
61 Before Netpbm 10.11 (October 2002), tifftopnm never used TIFFRGBAIm‐
63 ageGet(), so it could not interpret many of the formats it can inter‐
64 pret today.
65 There is no fundamental reason that this program could not read other
67 kinds of TIFF files even when they don't fit in memory all at once.
68 The existing limitations are mainly because no one has asked for more.
69 Output Image
72 The PNM output has the same maxval as the Tiff input, except that if
73 the Tiff input is colormapped (which implies a maxval of 65535) the PNM
74 output has a maxval of 255. Though this may result in lost informa‐
75 tion, such input images hardly ever actually have more color resolution
76 than a maxval of 255 provides and people often cannot deal with PNM
77 files that have maxval > 255. By contrast, a non-colormapped Tiff
78 image that doesn't need a maxval > 255 doesn't have a maxval > 255, so
79 when tifftopnm sees a non-colormapped maxval > 255, it takes it seri‐
80 ously and produces a matching output maxval.
81 Another exception is where the TIFF maxval is greater than 65535, which
83 is the maximum allowed by the Netpbm formats. In that case, tifftopnm
84 uses a maxval of 65535, and you lose some information in the conver‐
85 sion.
86
89 You may abbreviate any option to its shortest unique prefix. You may
90 use two hyphens instead of one in options. You may separate an option
91 and its value either by an equals sign or white space.
92 -alphaout=alpha-filename
96 tifftopnm creates a PGM file containing the alpha channel values
97 in the input image. If the input image doesn't contain an alpha
98 channel, the alpha-filename file contains all zero (transparent)
99 alpha values. If you don't specify -alphaout,
100 tifftopnm does not generate an alpha file, and if the input
102 image has an alpha channel, tifftopnm simply discards it.
103 If you specify - as the filename, tifftopnm writes the alpha
105 output to Standard Output and discards the image.
106 See pamcomp(1)foronewaytouse the alpha output file.
108 -respectfillorder
111 By default, tifftopnm ignores the 'fillorder' tag in the TIFF
112 input, which means it may incorrectly interpret the image. To
113 make it follow the spec, use this option. For a lengthy but
114 engaging discussion of why tifftopnm works this way and how to
115 use the -respectfillorder option, see the note on fillorder
116 below.
117 -byrow This option can make tifftopnm run faster.
120 tifftopnm has two different ways to do the conversion from Tiff
122 to PNM, using two different facilities of the TIFF library:
123 Whole Image
128 Decode the entire image into memory at once, using TIFFRGBAIm‐
129 ageGet(), then convert to PNM and output row by row.
130 Row By Row
133 Read, convert, and output one row at a time using TIFFReadScan‐
134 line().
135 Whole Image is preferable because the Tiff library does more of
139 the work, which means it understands more of the Tiff format
140 possibilities now and in the future. Also, some compressed TIFF
141 formats don't allow you to extract an individual row.
142 Row By Row uses far less memory, which means with large images,
144 it can run in environments where Whole Image cannot and may also
145 run faster. And because Netpbm code does more of the work, it's
146 possible that it can be more flexible or at least give better
147 diagnostic information if there's something wrong with the TIFF.
148 The Netpbm native code may do something correctly that the TIFF
150 library does incorrectly, or vice versa.
151 In Netpbm, we stress function over performance, so by default we
153 try Whole Image first, and if we can't get enough memory for the
154 decoded image or TIFFRGBAImageGet() fails, we fall back to Row
155 By Row. But if you specify the -byrow option, tifftopnm will
156 not attempt Whole Image. If Row By Row does not work, it simply
157 fails.
158 See Color Separation (CMYK) TIFFs ⟨#cmyk⟩ for a description of
160 one way Row By Row makes a significant difference in your
161 results.
162 Whole Image costs you precision when your TIFF image uses more
164 than 8 bits per sample. TIFFRGBAImageGet() converts the samples
165 to 8 bits. tifftopnm then scales them back to maxval 65535, but
166 the lower 8 bits of information is gone.
167 In many versions of the TIFF library, TIFFRGBAImageGet() does
169 not correctly interpret TIFF files in which the raster orienta‐
170 tion is column-major (i.e. a row of the raster is a column of
171 the image). With such a TIFF library and file, you must use
172 -byrow to get correct output.
173 Before Netpbm 10.11 (October 2002), tifftopnm always did Row By
175 Row. Netpbm 10.12 always tried Whole Image first. -byrow came
176 in with Netpbm 10.13 (January 2003).
177 -orientraw
180 A TIFF stream contains raster data which can be arranged in the
181 stream various ways. Most commonly, it is arranged by rows,
182 with the top row first, and the pixels left to right within each
183 row, but many other orientations are possible.
184 The common orientation is the same on the Netpbm formats use, so
186 tifftopnm can do its jobs quite efficiently when the TIFF raster
187 is oriented that way.
188 But if the TIFF raster is oriented any other way, it can take a
190 considerable amount of processing for tifftopnm to convert it to
191 Netpbm format.
192 -orientraw says to produce an output image that represents the
194 raw raster in the TIFF stream rather than the image the TIFF
195 stream is supposed to represent. In the output, the top left
196 corner corresponds to the start of the TIFF raster, the next
197 pixel to the right is the next pixel in the TIFF raster, etc.
198 tifftopnm can do this easily, but you don't get the right image
199 out. You can use pamflip to turn the output into the image the
200 TIFF stream represents (but if you do that, you pretty much lose
201 the benefit of -orientraw).
202 With this option, tifftopnm always uses the Row By Row method
204 (see -byrow).
205 This option was new in Netpbm 10.42 (March 2008). Before that,
207 tifftopnm generally produces arbitrary results with TIFF images
208 that have an orientation other than the common one.
209 -verbose
212 Print extra messages to Standard Error about the conversion.
213 -headerdump
216 Dump TIFF file information to stderr. This information may be
217 useful in debugging TIFF file conversion problems.
218
223 Fillorder
224 There is a piece of information in the header of a TIFF image called
225 'fillorder.' The TIFF specification quite clearly states that this
226 value tells the order in which bits are arranged in a byte in the
227 description of the image's pixels. There are two options, assuming
228 that the image has a format where more than one pixel can be repre‐
229 sented by a single byte: 1) the byte is filled from most significant
230 bit to least significant bit going left to right in the image; and 2)
231 the opposite.
232 However, there is confusion in the world as to the meaning of fil‐
234 lorder. Evidence shows that some people believe it has to do with byte
235 order when a single value is represented by two bytes.
236 These people cause TIFF images to be created that, while they use a
238 MSB-to-LSB fillorder, have a fillorder tag that says they used LSB-to-
239 MSB. A program that properly interprets a TIFF image will not end up
240 with the image that the author intended in this case.
241 For a long time, tifftopnm did not understand fillorder itself and
243 assumed the fillorder was MSB-to-LSB regardless of the fillorder tag in
244 the TIFF header. And as far as I know, there is no legitimate reason
245 to use a fillorder other than MSB-to-LSB. So users of tifftopnm were
246 happily using those TIFF images that had incorrect fillorder tags.
247 So that those users can continue to be happy, tifftopnm today continues
249 to ignore the fillorder tag unless you tell it not to. (It does, how‐
250 ever, warn you when the fillorder tag does not say MSB-to-LSB that the
251 tag is being ignored).
252 If for some reason you have a TIFF image that actually has LSB-to-MSB
254 fillorder, and its fillorder tag correctly indicates that, you must use
255 the -respectfillorder option on tifftopnm to get proper results.
256 Examples of incorrect TIFF images are at ftp://weather.noaa.gov.
258 ⟨ftp://weather.noaa.gov.⟩ They are apparently created by a program
259 called faxtotiff.
260 This note was written on January 1, 2002.
262 Color Separation (CMYK) TIFFs
266 Some TIFF images contain color information in CMYK form, whereas PNM
267 images use RGB. There are various formulas for converting between
268 these two forms, and tifftopnm can use either of two.
269 The TIFF library (Version 3.5.4 from libtiff.org) uses Y=(1-K)*(1-B)
271 (similar for R and G) in its TIFFRGBAImageGet() service. When
272 tifftopnm works in Whole Image mode, it uses that service, so that's
273 the conversion you get.
274 But when tifftopnm runs in Row By Row mode, it does not use TIFFRGBAIm‐
276 ageGet(), and you get what appears to be more useful: Y=1-(B+K). This
277 is the inverse of what pnmtotiffcmyk does.
278 See the -byrow option for more information on Whole Image versus Row By
280 Row mode.
281 Before Netpbm 10.21 (March 2004), tifftopnm used the Y=(1-K)*(1-B) for‐
283 mula always.
284
288 pnmtotiff(1), pnmtotiffcmyk(1), pamcomp(1), pnm(1)
289
292 Derived by Jef Poskanzer from tif2ras.c, which is Copyright (c) 1990 by
293 Sun Microsystems, Inc. Author: Patrick J. Naughton
294 (naughton@wind.sun.com).
295netpbm documentation 12 July 2009 Tifftopnm User Manual(0)