![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to cmd.ms-1 CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Research Unix] / researchv10dc / vol2 / graphics|
Return to cmd.ms-1 CVS log | Up to [Research Unix] / researchv10dc / vol2 / graphics |
1.1 root 1: .so ../ADM/mac
2: .XX 30 491 "The 10th Edition Raster Graphics System"
3: .fp 5 T CW \" T for Typewriter
4: .de TP \" An indented paragraph describing some command, tagged with the command name
5: .IP "\\fT\\$1\\fR" 8
6: .if \\w'\\fT\\$1\\fR'-7n .br
7: ..
8: .de CI
9: .nr Sf \\n(.f
10: \%\&\\$3\f(CW\\$1\fI\&\\$2\f\\n(Sf
11: ..
12: .TL
13: The 10th Edition Raster Graphics System
14: .AU
15: Tom Duff
16: .AI
17: .MH
18: .AB
19: The current (late 1989) state of generating and displaying raster graphics
20: in Research
21: .UX
22: is described.
23: .AE
24: .2C
25: .NH
26: Introduction
27: .PP
28: The Research
29: .UX
30: system contains a number of commands to capture, manipulate,
31: display and record monochrome and full-color raster images. Three groups of
32: commands may be identified:
33: interactive programs that operate on a frame-buffer,
34: commands that operate on images stored in picture
35: files (see
36: .I picfile (5)
37: or |reference(picfile v10)),
38: and programs that interface to various graphical I/O devices:
39: video cameras, scanners, paper plotters, film cameras and video-tape
40: recorders.
41: .NH
42: Video Facilities
43: .PP
44: No discussion of our raster graphics software can ignore the
45: hardware on which it runs. The hardware available at different
46: sites will, of course, vary. For definiteness, and to provide
47: help for the local audience, this section will discuss the
48: hardware available in Center 1127's graphics and image processing
49: laboratory (MH 2C-524) and its neighborhood. Most other environments
50: will have hardware that is similar in spirit if different in detail.
51: .PP
52: There are seven work stations in 2C-528. On the day this was written,
53: five of them had TTY 5620 terminals, one had a Gnot terminal and
54: one had a SUN-3 workstation computer. Eventually, most of the 5620's
55: will be replaced with Gnots. Each work station also has
56: a Sony GDM-1901-12 video monitor that displays high-resolution
57: video signals.
58: .PP
59: The room contains other video displays and recorders, including
60: a Barco video projector in the ceiling, a 35-inch Mitsubishi monitor
61: at the front of the room, a 19-inch Barco monitor at work station 6,
62: two small Sony monitors in the video rack next to the audio console,
63: two Panasonic Super-VHS recorders, a Sony 3/4-inch (U-MATIC) video
64: player, a multi-standard (SECAM, NTSC, PAL) VHS player, a
65: Sony BVH-2500 1-inch (SMPTE-C) video tape recorder and a Sony video
66: camera.
67: .PP
68: The video equipment supports at least three incompatible video formats.
69: High-resolution
70: rgb video has 1024 scan lines, a 60 hz non-interlaced vertical scan rate,
71: and transmits red, green and blue information on separate cables with
72: synchronization pulses superimposed on the green channel.
73: Low-resolution rgb video has between 480 and 488 scan lines, 30hz
74: interlaced vertical scan, and separate rgb with sync on green.
75: NTSC (National Television Standards Committee) video has the same timing
76: characteristics as low-resolution rgb video, but encodes red, green, blue
77: and sync into a single signal.
78: NTSC is the encoding used by American, Canadian and Japanese television
79: broadcasters, and by almost all video recording and playback equipment
80: in those countries.
81: .PP
82: Various computer terminals generate video in other formats that our
83: equipment handles with only limited success. Gnots, 630s,
84: 5620s, Sun terminals, IBM-compatible PCs and Macintoshes all generate
85: mutually incompatible video. Their vertical and horizontal scan-rates
86: differ. The voltages and impedences of the signals they produce differ.
87: Their color encodings differ. Monitors that can display video from all
88: of these sources are rare, let alone hardware to convert from one format
89: to another. For example, the only reliable way to record a signal from
90: any of these sources is to place a camera in front of a monitor. The
91: quality of the resulting recordings is often bad. It is a black art
92: to adjust our Barco video projector to handle non-standard signals,
93: but with a few days notice it can often be done. Again, the results
94: are not often as good as one might like \(mi the projector does not
95: focus as tightly as a monitor and its brightness is limited. As better
96: video displays become available our situation will improve.
97: Table 1 summarizes the equipment available and the
98: video formats that each supports.
99: .1C
100: .KF
101: .TS
102: center box;
103: l| c| c| c| c| c| c.
104: Equipment High-res Low-res NTSC Gnot IBM PAL/SECAM
105: RGB RGB
106: =
107: workstation monitors \(bu
108: _
109: Barco projector \(bu \(bu \(bu \(bu maybe maybe
110: _
111: 35-inch Mitsubishi \(bu \(bu \(bu
112: _
113: Barco at work station 6 \(bu \(bu
114: _
115: Sony rack monitors \(bu
116: _
117: Super-VHS recorders \(bu
118: _
119: U-MATIC player \(bu
120: _
121: 1-inch recorder \(bu
122: _
123: multi-standard player \(bu
124: _
125: camera \(bu \(bu
126: _
127: Metheus frame buffers \(bu
128: _
129: ITI frame buffer \(bu
130: _
131: Pixel Machine \(bu \(bu
132: .TE
133: .sp .3
134: .ce
135: \fBTable 1. \fRVideo devices
136: .SP .5
137: .KE
138: .2C
139: .PP
140: We have several Metheus 3610 frame buffers (four on
141: .CW pipe ,
142: one on
143: .CW arend ,
144: one on
145: .CW encke )
146: and an Imaging Technology, Inc. (ITI) RGB-512. All of our
147: frame buffers store 32 bits at each pixel, one byte each for red, green
148: blue and alpha. The 3610's generate high-resolution (1280\(mu1024)
149: video. The ITI generates low-resolution (512\(mu480) video that may
150: be recorded on video tape after conversion to NTSC. Connected to
151: .CW sol ,
152: the Sun-3/160 at work station 1, is an AT&T Pixel Machine with
153: 58 processors. It can generate either high- or low-resolution video
154: under software control; the Pixel Machine documentation can tell you how.
155: .PP
156: Each piece of video equipment is connected to one of two
157: video switches. One switch handles high or low-resolution rgb video.
158: The other handles NTSC video exclusively. Under computer control, any
159: output port may be connected to any input port of the same switch.
160: The
161: .CW "vsw [-dcrvpi] [\fIiopair\fP ...]
162: command controls the video switches.
163: The switches' inputs and outputs are labelled by single characters
164: 0-9 and a-z.
165: An
166: .I iopair
167: is a pair of characters indicating an input and output to be connected
168: together.
169: The flags mean:
170: .TP -r
171: operate the rgb switch.
172: .TP -c
173: operate the composite (NTSC) switch.
174: .TP -d
175: print a description of the switch's inputs and outputs
176: .TP -p
177: print a list of iopairs indicating the state of the switch.
178: .TP -i
179: initialize the switch to an unknown state.
180: .TP -v
181: with
182: .CW -p
183: or
184: .CW -d ,
185: make the descriptions more verbose.
186: .LP
187: The video switch is
188: .I not
189: straightforward to use.
190: It is easy to be confused about what the inputs and outputs are;
191: always remember that these are from the switch's point of view.
192: Thus, the input marked
193: .CW "pixel machine"
194: is the Pixel Machine's output.
195: The connections to the video switch have a history of ad hoc undocumented
196: changes.
197: .PP
198: The Sony BVH-2500 video recorder produces very high
199: quality recordings on 1" video tape. Since it can overwrite
200: arbitrary single frames of the tape, it is an ideal machine
201: on which to record animation.
202: .PP
203: Before using a new tape,
204: you must record (``grind'') time-code on it, numbering
205: each frame of the tape.
206: Time-code values are usually denoted by values
207: of the form
208: .I hh.mm.ss.ff
209: (like
210: .CW 02.43.17.15 ).
211: The 2500's monitor output, called
212: .CW 2500.mon
213: by
214: .I vsw ,
215: displays time-code superimposed on the 2500's
216: output signal.
217: .PP
218: To grind time-code, use
219: .I vsw
220: to connect the color bar generator
221: (\fIvsw\fP calls this
222: .CW cbars )
223: to the 2500's input, thread up a tape, and manually set the
224: 2500 to record by pushing its
225: .I record
226: and
227: .I play
228: buttons simultaneously. Let it go until the tape runs out.
229: .PP
230: The
231: .I 2500
232: command operates the recorder, reading instructions
233: from its standard input. Its instruction set is
234: moderately complicated; for most uses the following
235: subset is adequate:
236: .TP "cue \fIhh.mm.ss.ff
237: Cue the tape to the given time code. The
238: time-code displayed on the 2500's monitor
239: output may be a few frames off,
240: but the recorder will be cued to the correct point.
241: .TP "still mode on
242: Put the recorder in single-frame record mode.
243: .TP "still mode off\fP
244: Put the recorder out of single-frame record mode.
245: .TP "snap [\fIn\fT]
246: Record
247: .I n
248: frames (default 1) at the current cue point, and
249: advance the cue point by
250: .I n
251: frames. The recorder must be in single frame mode.
252: .TP "play
253: Start playing back from the current cue point.
254: .TP "stop
255: Stop the recorder.
256: .TP "!\fIunix-command
257: Run the given
258: .I unix-command
259: using
260: .CW /bin/sh .
261: .PP
262: We currently have only two sources of digital video that
263: may be recorded on video tape. These are the Pixel Machine
264: and the ITI frame buffer attached to kwee. To use either
265: one, you must use
266: .CW "vsw -r
267: to patch its output to the NTSC color encoder,
268: and
269: .CW "vsw -c
270: to patch the endcoder's output to the video recorder.
271: The ITI frame buffer is also useful as a frame-grabber,
272: capturing its video input in its memory whence it may
273: be saved in a picture file or otherwise manipulated.
274: .PP
275: The ITI is served by an ancient software regime whose
276: commands all begin with the letters
277: .CW iti .
278: .TP "itifbinit -x
279: Re-initialize the ITI to the state expected by the
280: rest of the software. The ITI is often unused for
281: days at at time, during which its health often
282: decays.
283: .CW itifbinit
284: is its restorative. The
285: .CW -x
286: flag causes its input and output to be synchronized.
287: This is always a good idea.
288: .TP itigamma
289: Load the ITI's color map to correct intensities
290: for display on CRT monitors.
291: .TP "itigrab -[gs]
292: Run the frame-grabber. The
293: .CW -g
294: flag starts the frame-grabber running.
295: The displayed image will track the ITI's input
296: video.
297: .CW -s
298: stops the frame-grabber, freezing the image.
299: Unadorned by flags,
300: .CW grab
301: starts the frame-grabber and stops it one frame
302: later.
303: .TP "itigit \fIpicture-file
304: Copy the image stored in
305: .I picture-file
306: into the ITI.
307: .TP "itisiv \fIpicture-file
308: Save the image in the ITI in
309: .I picture-file .
310: .PP
311: Paragraphs needed describing imagitek scanner, henry's document
312: scanners, matrix qcr, postscript.
313: .NH
314: Frame-buffer commands
315: .PP
316: A frame buffer is a large memory organized as a two-dimensional array of
317: pixels. Our Metheus 3610 frame buffers have 1024 scan lines of 1280 pixels
318: each. The ITI frame buffer has 480 lines of 512 pixels. The coordinate
319: system has (0,0) in the upper left-hand corner, with x increasing to
320: the right, and y increasing down. This apparent weirdness is fairly standard,
321: since it makes video output happen in row-major order.
322: .PP
323: Here we will mostly discuss commands for the Metheus displays. The corresponding
324: ITI commands have the same names, but prefixed with the string
325: .CW iti .
326: .PP
327: There are several Metheus frame buffers attached to pipe, named
328: .CW /dev/om[0-7] .
329: All of the commands discussed below determine which one to use
330: by examining the environment variable
331: .CW FB .
332: It is often hard to tell what frame buffer is displayed on which monitor
333: because of the video switch. The
334: .CW fbi
335: (frame buffer identification) command displays each frame buffer's name in it.
336: .PP
337: Our frame buffers all have 32 bits per pixel, divided into
338: four 8-bit channels. The channel values are normally thought of
339: as fractions ranging from 0 to 1, although frame buffer commands perversely
340: refer to them as integers between 0 and 255.
341: Three of the channels specify the red, green and
342: blue color components of the image. The fourth channel, called
343: .I alpha ,
344: is used to indicate whether or not the image covers the pixel, and is not
345: normally displayed.
346: .I Alpha
347: is used to control image compositing operations|reference(porter duff compositing digital images).
348: Fractional values of
349: .I alpha
350: describe pixels that the image partly or translucently covers, and
351: facilitate anti-aliased compositing.
352: .PP
353: Each frame buffer contains three 256 entry look-up tables that specify
354: mappings from the values stored in the red, green and blue channels
355: to the voltages supplied at the frame buffers' video outputs.
356: The
357: .CW "gamma [\fIpower\fP]
358: command loads these tables are loaded with a function that
359: inverts the power-law relation between voltage and luminous flux
360: normally encountered in CRT displays. Thus, pixel values normally
361: correspond directly to displayed intensities.
362: .I Power
363: is the exponent of the power-law. The default of 2.3 is adequate
364: for all our displays.
365: .PP
366: Other patterns may be loaded into the color map by
367: .CW "getmap \fIfile\fP [...]
368: command, whose arguments are a list of files containing color maps.
369: On the ITI, the argument `\fT%\fP' refers
370: to the current content of the frame buffer's color map. (The Metheuses'
371: color maps are write-only.) The functional composition of the specified
372: color maps is loaded into the frame buffer's color map.
373: .CW Getmap
374: searches for files in the current directory and in
375: .CW /fb/cmap
376: and
377: .CW /usr/td/cmap/lib .
378: A color map file contains 256 records of 3 bytes each, specifying the output
379: values for the corresponding red, green and blue input values.
380: The
381: .CW ranmap
382: command loads random values into the color map.
383: .PP
384: Our frame buffers have the ability to magnify a part of the image, via the
385: .CW "zoom [\fIamount\fP [\fIx y\fP]]
386: command.
387: With three arguments,
388: .CW zoom
389: magnifies by
390: .I amount ,
391: mapping the point
392: .I "(x, y)
393: (default (0,0)) to the upper left-hand corner of the screen.
394: With no arguments,
395: .I Amount
396: defaults to 1.
397: The Metheuses can magnify by any integral factor from 1 to 16.
398: The ITI can magnify only by 1 or 2.
399: .PP
400: An array of images can be viewed in sequence by zooming and panning
401: using the
402: .CW "movie \fIxsize ysize nx ny\fP [\fIdelay\fP]
403: command. The arguments are the size of the individual
404: frames, the number of frames in the array in each
405: direction, and optionally the number of 60ths of a second
406: to delay between frames. The frames must be arranged
407: boustrophedonically, with alternate rows proceding from
408: left to right and right to left. (This is because neither
409: Metheus nor ITI frame buffers can pan in x and y simultaneously
410: without glitching.)
411: .PP
412: There are a number of commands to load simple patterns into the frame buffer:
413: .TP "clr [-w \fIx0 y0 x1 y1\fT] [\fIr\fT [\fIg b\fT [\fIalpha\fT]]]
414: sets all pixels to the given value. If only
415: .I r
416: is given,
417: .I g
418: and
419: .I b
420: are set to
421: .I r .
422: If
423: .I alpha
424: is not given, it is set to 255 (completely opaque.)
425: The
426: .CW -w
427: flag restricts attention to pixels inside the window whose upper-left corner is
428: .I (x0,y0)
429: and with
430: .I (x1,y1)
431: just diagonally outside the lower-right corner.
432: .TP cbars
433: displays a color-bars test pattern. The 8 bars at the top exercise all combinations
434: of the 3 primary colors. The 9 patches at the bottom are a logarithmic
435: (perceptually uniform) grey scale.
436: .TP "ramp [-w \fIx0 y0 x1 y1\fT] [-v] [[\fIcol0\fT] \fIcol1\fT]
437: displays a horizontal ramp whose color is
438: .I col0
439: at the left
440: and
441: .I col1
442: at the right.
443: Colors are specified as for
444: .CW clr
445: (green and blue default equal to red, alpha defaults to 255).
446: If
447: .I col0
448: defaults to
449: .CW "0 0 0 255" .
450: .CW -w
451: restricts
452: .CW clr
453: to the given window.
454: .CW -v
455: gives a vertical ramp with
456: .I col0
457: at the top
458: and
459: .I col1
460: at the bottom.
461: .TP "colors [-gfr]
462: displays a 16 by 16 array of grey-colored (equal red, green and blue) squares in
463: the middle of the screen with red, green and blue ramps at the top. This is
464: mostly useful for examining color maps. The flags modify the display in small ways.
465: .CW -r
466: suppresses the ramps.
467: .CW -g
468: suppresses the gaps between the squares.
469: .CW -f
470: expands the display to fill the full screen, making the patches non-square and
471: suppressing the ramps.
472: .PP
473: The
474: .CW xhair
475: command can be used to examine the contents of the frame buffer.
476: It is named after the cross-hair that it draws on the screen. Single
477: character commands manipulate the cross-hair, magnify the video and
478: print pixel values. The commands are
479: .nf
480: .ta 8n
481: \fTh\fP print the help message
482: \fTlrud\fP move left, right, up or down 1 pixel
483: \fTLRUD\fP move left, right, up or down 16 pixels
484: \fT0\fP move to center of screen (x=256, y=240)
485: \fT1-8\fP magnify \(mu1\-8
486: \fT9\fP magnify \(mu16
487: \fTp\fP print current coordinates and pixel value
488: \fTP\fP print pixel after each command (toggle)
489: \fTm\fP type coordinates to move to
490: \fTx\fP type x coordinate to move to
491: \fTy\fP type y coordinate to move to
492: \fTc\fP change the crosshair display to a rectangle
493: \fTs\fP manipulate other corner of rectangle
494: \fT^D,q\fP exit xhair and run command
495: \fTQ\fP exit xhair, don't demagnify or run command
496: \fTX\fP exit and don't run command
497: .PP
498: If
499: .CW xhair
500: is given arguments, they represent a command to be executed before exiting,
501: after making substitutions for any argument whose first character is
502: .CW % .
503: The substitutions made are:
504: .ta 8n
505: .nf
506: \fT%r\fP the current rectangle
507: \fT%w\fP the current rectangle
508: \fT%p\fP the upper-left corner of the rectangle
509: \fT%o\fP the upper-left corner of the rectangle
510: \fT%c\fP the lower-right corner of the rectangle
511: \fT%x\fP the x coordinate of the upper-left corner
512: \fT%y\fP the y coordinate of the upper-left corner
513: \fT%X\fP the x coordinate of the lower-right corner
514: \fT%Y\fP the y coordinate of the lower-right corner
515: .NH
516: Picture file commands
517: .PP
518: Most of our raster graphics commands require no special hardware. They synthesize
519: images in picture files from textual or other discriptions, they modify images
520: in picture files, producting results in picture files, or they combine the contents
521: of several picture files to produce composite images, again storing the result in
522: a picture file.
523: .PP
524: The simplest picture file command is
525: .CW pcp ,
526: which takes two picture file names and copies the first onto the second.
527: Names whose first character is
528: .CW %
529: are interpreted specially:
530: .ta 8n
531: .nf
532: \fT%in\fP read from standard input
533: \fT%out\fP write standard output
534: \fT%0\fP Metheus frame buffer #0.
535: \&...
536: \fT%7\fP Metheus frame buffer #7.
537: \fT%iti\fP ITI frame buffer.
538: .fi
539: .PP
540: .CW Pcp
541: has a number of options that alter the copied picture:
542: .TP "-o\ \fIx y
543: Add
544: .I "(x,y)
545: to the picture's window coordinates.
546: .TP "-w \fIx0\ y0\ x1\ y1
547: Clip the input picture's window to the given coordinates.
548: If
549: .CW -o
550: and
551: .CW -w
552: are both given, the window is clipped before being offset.
553: .TP "-t\ \fItype
554: The output picture will have
555: .CW "TYPE= \fItype" .
556: .TP "-c\ \fIchannels
557: The output picture will have
558: .CW "CHAN=\fIchannels" .
559: .I Channels
560: must be one of
561: .CW m ,
562: .CW ma ,
563: .CW rgb ,
564: .CW rgba ,
565: .CW mz... ,
566: .CW maz... ,
567: .CW rgbz... ,
568: or
569: .CW rgbaz... .
570: Likewise, the input picture must
571: have
572: .CW CHAN=
573: set to one of these possibilities.
574: If the input has no
575: .CW a
576: channel, 255 is used.
577: If it has no
578: .CW z...
579: channels, floating point 0. is used.
580: Color images are converted to monochrome using the standard
581: NTSC luminance (\fTa=.30r+.59g+.11b\fP).
582: Monochrome images are converted to color by \fTr\fP=\fTg\fP=\fTb\fP=\fTm\fP.
583: .PP
584: The
585: .CW lam
586: command combines any number of images, writing a picture file whose window
587: is large enough to contain all the windows of its inputs. The input files
588: are combined with pixels of later images overwriting earlier ones. As
589: this is only really useful if the windows of the input images differ,
590: each input name may be preceded by
591: .CI -w "x0 y0 x1 y1
592: and
593: .CI -o " x y
594: options to clip and offset it.
595: Options
596: .CI -W " x0 y0 x1 y1
597: and
598: .CI -O " x y
599: clip and offset the output image.
600: Option
601: .CI -o " file
602: specifies the output file name (standard output by default),
603: and
604: .CI -c " channels
605: specifies the output
606: .CW CHAN= .
607: .PP
608: The
609: .CW posit
610: and
611: .CW 3matte
612: commands combine images using the two- and three-dimensional compositing
613: operations described in |reference(porter duff compositing digital images)
614: and |reference(duff composite3d).
615: Each takes a list of picture file names as arguments, producing a composite
616: on standard output. The
617: .CW -a
618: option will cause either program to output only the
619: .CW rgb
620: channels, suppressing
621: .CW a
622: (and
623: .CW z...
624: in the case of
625: .CW 3matte ).
626: .PP
627: There is an army of commands to read an image and, under the control of a few
628: parameters, write a modified image. These include:
629: .TP "bwquantize \fIin out
630: File
631: .I in
632: contains a color image or a monochrome image with a colormap.
633: A gray-level image is written in
634: .I out ,
635: using the NTSC luminance formula.
636: .TP "clip \fIleft right in out
637: Clip an image horizontally to the limits
638: .I left
639: and
640: .I right .
641: A picture that does not fill out the limits is filled with black pixels.
642: .TP "xpand [-s] \fIin out [lo hi [inlo inhi]]
643: The input picture has its dynamic range adjusted so that pixels in the
644: range
645: .I inlo
646: to
647: inhi
648: are mapped to the range
649: .I lo
650: to
651: .I hi
652: (default 0 to 255).
653: The default values for
654: .I inlo
655: and
656: .I inhi
657: are determined per channel by examining the input picture.
658: The
659: .CW -s
660: option causes all channels to be examined together.
661: .I Lo ,
662: .I hi ,
663: .I inlo
664: and
665: .I inhi
666: may have any values whatsoever. If
667: .I hi
668: is smaller than
669: .I lo ,
670: pixel values will be inverted, producing a negative image.
671: Any output pixel that would be mapped outside the range
672: 0\-255 is set to 0 or 255.
673: .TP "dither \fIin out
674: Convert a full-color (3 channel) picture to one channel with a color-map
675: by dithering.
676: .TP "floyd \fIin out
677: Convert an 8-bit gray-scale picture to one bit per pixel using a version of the Floyd-Steinberg
678: error-diffusion method.
679: .TP "halftone \fIscreen in out
680: Convert an 8-bit gray-scale picture to one bit using a given half-tone
681: .I screen .
682: The available screens are
683: .KS
684: .in 2n
685: .TS
686: lFCW l.
687: ALLEBACH Allebach's ordered-dither
688: BAYER Standard ordered-dither
689: BLUENOISE A pebble-screen pattern
690: CLASSIC A 3-pixel-wide dot screen
691: CLASSIC2 Another 3-pixel-wide dot screen
692: CLASSIC3 A 4-pixel-wide dot screen
693: CLASSIC4 An 8-pixel-wide dot screen
694: DIAMOND Rao and Arce's ordered-dither
695: LINE Ulichney's line screen
696: RING A concentric ring screen
697: TILT18 A tilted dot screen
698: .TE
699: .KE
700: .TP "he \fIin out
701: Histogram equalization: the intensity histogram
702: of the input image is measured. The output image
703: has its contrast altered for maximum use of the
704: output range, equalizing the histogram
705: as much as possible.
706: .TP "hysteresis \fIlow high in out
707: Pixel values of the input picture below
708: .I low
709: are mapped to zero.
710: Those above
711: .I high
712: are mapped to 255.
713: If
714: .I low
715: and
716: .I high
717: are not equal, any region
718: below
719: .I high
720: that has any 8-connected neighbors below
721: .I low
722: is mapped to zero.
723: .TP "picaverage \fIweight in1 in2 out
724: The output picture is a weighted average of the two
725: input pictures.
726: .I Weight
727: determines the fraction of the average contributed by
728: .I in1 .
729: .TP "piccat \fTin1 ... out
730: The input pictures are concatenated one atop another. The
731: output has the width of the widest input.
732: .TP "picjoin \fTin1 ... out
733: The input pictures are concatenated side by side. The
734: output has the height of the highest input.
735: .TP "adapt \fIin out
736: Adaptive histogram equalization: each pixel of the output image
737: is the histogram-equalized value of the center of a 7\(mu7 pixel
738: window surrouding it in the input image.
739: .TP "ahe \fIin out
740: 17\(mu17 adaptive histogram equalization. There should be an option to
741: .CW adapt
742: setting the window size.
743: .TP "clean \fIin out
744: Bayer-Powell noise removal filter. If the center pixel of each 3\(mu3
745: window in the input differs from the average of the other 8 pixels by
746: more than 64, it is replaced by the periphery-average. This has the
747: effect of flattening isolated noise pixels.
748: .TP "crispen \fIin out
749: 3\(mu3 linear crispening filter. Convolves the input image with
750: the kernel
751: .P1
752: -1 -1 -1
753: -1 9 -1
754: -1 -1 -1
755: .P2
756: This is a mild high-pass filter.
757: .TP "edge \fIin out
758: 3\(mu3 linear edge-detection filter.
759: Convolves the input image with
760: the kernel
761: .P1
762: -1 -1 -1
763: -1 8 -1
764: -1 -1 -1
765: .P2
766: This is just the difference between the original image and the output of
767: .CW crispen .
768: .TP "edge2 \fIin out
769: 3\(mu3 non-linear edge-detection (Sobel operator) filter.
770: .TP "extremum \fIin out
771: 3\(mu3 extremum filter. Replaces the center pixel of each
772: by the value in the 3\(mu3 window surrounding it that most
773: differs from it.
774: .TP "laplace \fIin out
775: 3\(mu3 "Laplacian filter.
776: Convolves the input image with
777: the kernel
778: .P1
779: 0 -1 0
780: -1 5 -1
781: 0 -1 0
782: .P2
783: This is a fairly extreme high-pass filter.
784: .TP "median \fIin out
785: 3\(mu3 median filter. Each pixel is replaced by the
786: median of the 3\(mu3 window surrounding it.
787: .TP "smooth \fIin out
788: 3\(mu3 Bartlett filter.
789: Convolves the input image with
790: the kernel
791: .P1
792: 1/16 2/16 1/16
793: 2/16 4/16 2/16
794: 1/16 2/16 1/16
795: .P2
796: This is a moderately strong low-pass filter.
797: .TP "picreflect \fIin out
798: Reflect the input picture about its vertical center line.
799: .TP "quantize \fIin out
800: Convert a full-color picture to an 8-bit picture with colormap.
801: This does a much better job than
802: .CW dither .
803: It tries to pick an colormap that optimally clusters the input's
804: colors.
805: .TP "resample \fInpix in out
806: Resample the input image to be
807: .I npix
808: pixels wide. The filter used in resampling minimizes both
809: pre- and post-aliasing.
810: .TP "transpose \fIinput output
811: Transpose the input picture. This is useful in conjunction with
812: commands that operate on scan-lines, like
813: .CW resample
814: or
815: .CW clip ,
816: to perform operations on columns instead of rows.
817: .TP "shear \fIin out angle
818: Rotate the input image by the given
819: .I angle
820: (in degrees). It's called
821: .I shear
822: because it operates by shearing the image 3 times
823: (horizontally, then vertically, then horizontally).
824: It fails for angles of 180\(de.
825: .TP "lx [-o\fIfile\fT] [-A\fIaspect\fT] [-a] [-s\fIscale\fT] [-r\fIrot\fT] [-x\fIxscale\fT] [-y\fIyscale\fT] \fIin\fT
826: Perform a linear transformation on the input image.
827: The
828: .CW -o
829: option specifies the output file name. The default is standard output.
830: The
831: .CW -A
832: option specifies the aspect ratio of the pixels. The default is 1.
833: The ITI frame-grabber produces images whose pixel aspect-ratio is 1.25.
834: The
835: .CW -a
836: option suppresses the writing of an alpha channel. Normally
837: an alpha channel is computed even for input images that don't have one,
838: since the output picture is often rotated and thus doesn't completely
839: cover its window.
840: .IP
841: The transformation is specified by a sequence of options. The specified
842: transformations are combined in the order given to yield a composite transformation.
843: The relevant options are:
844: .nf
845: .ta 8n
846: \fT-s\fIscale\fR scale by \fIscale\fR.
847: \fT-r\fIrot\fR rotate by \fIrot\fR degrees clockwise.
848: \fT-x\fIxscale\fR scale in x by \fIxscale\fR.
849: \fT-y\fIyscale\fR scale in y by \fIyscale\fR.
850: .PP
851: There are several commands to generate images from
852: three-dimensional geometric descriptions of various sorts.
853: Most of these produce
854: .CW CHAN=rgbaz...
855: images that may be combined using
856: .CW 3matte .
857: In their output files, points at the near clipping plane will be mapped to
858: points having
859: .I z=0 ,
860: and points at the far clipping plane will have
861: .I z=1 .
862: .TP "ncpr [-a \fIaspect\fT] [-w \fIx0 y0 x1 y1\fT] [-c \fIrgbaz\fT] \fIinput\fT [\fIoutput\fT]
863: New Cheezy Polygon Renderer.
864: .I Output
865: (default standard output) is the name of the picture file that will contain the rendered
866: version of the scene described in
867: .I input ,
868: a text file specifying a polygonal scene.
869: The
870: .CW -a
871: option sets the pixel aspect-ratio (default 1.)
872: The
873: .CW -w
874: option sets the window of the output picture.
875: The
876: .CW -c
877: option specifies which channels should be written to the output picture.
878: .IP
879: The input file contains a sequence of single-letter commands, each with several
880: numeric parameters. The commands are:
881: .IP
882: \fTv \fIfov near far ex ey ez lx ly lz ux uy uz\fR
883: .br
884: Set viewing parameters.
885: .I Fov
886: is the angle subtended vertically by the screen at the eye point.
887: Points whose distance from the eye is not between
888: .I near
889: and
890: .I far
891: will be clipped away before drawing. However tempted, do not set
892: .I near
893: to zero, lest underflow or divide-check occur.
894: .I "(ex, ey, ez)
895: is the coordinate of the eye, the point from which the scene is viewed
896: and the center of perspective.
897: .I "(lx, ly, lz)
898: is a vector pointing from the eye toward the center of the scene.
899: The point
900: .I "(lx+ex, ly+ey, lz+ez)
901: is mapped into the center of the screen.
902: .I "(ux, uy, uz)
903: is the up vector, the direction of the zenith. The point
904: .I "(lx+ux, ly+uy, lz+uz)
905: is mapped into a point somewhere above the center of the screen.
906: .IP
907: \fTl \fIx y z\fR
908: .br
909: Set the direction of the light source to
910: .I "(x, y, z) .
911: The light source is ``at infinity'' in the given direction.
912: .IP
913: \fTb \fIred green blue alpha\fR
914: Clear the screen to the given color.
915: .I Red ,
916: .I green ,
917: .I blue
918: and
919: .I alpha
920: should all be between 0 and 255.
921: .IP
922: \fTc \fIindex red green blue alpha\fR
923: Set a color table entry. Indices into the color table are used
924: to specify the colors of polygons (see below.)
925: The table has 500 entries. Unless reloaded by the
926: .CW c
927: command, the first 256 entries contain the 256 shades of gray,
928: the following 12 entries (256-267) are set to 12 logarithmically
929: spaced (perceptually equal) gray shades, and the next 20 entries
930: (268-287) to 20 logarithmically spaced gray shades.
931: .IP
932: \fTt \fIx0 y0 z0 x1 y1 z1 x2 y2 z2 c0 c1\fR
933: .br
934: Render a triangle with vertices
935: .I "(x0, y0, z0)" ,
936: .I "(x1, y1, z1)
937: and
938: .I "(x2, y2, z2)" .
939: On one side its color is
940: .I c0 ,
941: on the other it is
942: .I c1 .
943: If
944: .I c0
945: or
946: .I c1
947: is positive, the polygon's color is found in the corresponding color
948: table entry. If negative, the color is found by modifying the color
949: table entry as though the surface were illuminated by a light source
950: whose direction was specified by the
951: .CW l
952: command.
953: .IP
954: \fTp \fIc0 c1 x0 y0 z0 x1 y1 z1 ... xn yn zn \fT;\fR
955: .br
956: Render a polygon whose color is
957: .I c0
958: on one side and
959: .I c1
960: on the other.
961: The polygon's vertices are
962: .I "(x0, y0, z0)" ,
963: .I "(x1, y1, z1)" ,...,
964: .I "(xn, yn, zn)" .
965: .in -8n
966: .TP "quad [-a] [-z] [-w \fIx0 y0 x1 y1\fP] \fIin out
967: .br
968: Compute an image of a quadric surface. The
969: .CW -a
970: option suppresses writing out the alpha channel.
971: The
972: .CW -z
973: option suppresses writing out the z channel.
974: The
975: .CW -w
976: option specifies the output window.
977: The input file should contain 34 floating point numbers.
978: The first ten numbers are the upper triangle of the symmetric
979: matrix describing the quadratic form (in screen coordinates.)
980: The next 16 numbers are a matrix that converts screen-space
981: coordinates into world-space normals for illumination computations.
982: The next three numbers are the direction of the light source.
983: The next four numbers are the red, green, blue and alpha of the
984: surface's color. The last number is the amount of ambient light
985: in the environment.
986: .TP "terrain \fIin out ex ey ez lx ly fov near far
987: Render a terrain image.
988: The input file
989: is a 2-channel picture file containing 16-bit elevation data on a regular grid.
990: .I "(ex, ey, ez)
991: is the eye position.
992: .I "(lx, ly, 0)
993: is a vector pointing from the eye to the center of the scene.
994: The up direction is
995: .I "(0,0,1)" .
996: .I Fov
997: is the vertical field-of-view angle.
998: .I Near
999: and
1000: .I far
1001: are the distances from the eye to the near and far clipping planes.
1002: .TP "bg \fIr0 g0 b0 r1 g1 b1 out
1003: Generate a background card whose color varies smoothly
1004: from
1005: .I "(r0, g0, b0)
1006: at the top to
1007: .I "(r1, g1, b1)
1008: at the top.
1009: Its z coordinate is set to 2, which is beyond the far clipping plane.
1010: .NH
1011: Animation
1012: .PP
1013: To use a command-based raster graphics system as described here to
1014: for animation requires writing command files to create and record
1015: long sequences of images. Typical command files contain long sequences
1016: of repeated commands with slowly changing numeric parameters. Several
1017: sequences starting and ending at different times may be interleaved
1018: to describe overlapping motion. They are at best tedious and at worst
1019: tricky to generate by hand or using the usual tools.
1020: .PP
1021: .I Moto
1022: is a command generator tailored for an animator's needs. Its input is
1023: a concise description of the animation to be performed; its output is
1024: a command file suitable for input to
1025: .I sh ,
1026: .I rc
1027: or some other command interpreter.
1028: Its arguments are an optional file name containing a
1029: .I moto
1030: program (default standard input) and list of numeric parameters
1031: that are made available to the program.
1032: .PP
1033: A
1034: .I moto
1035: program consists of a list of groups of commands. Each block is
1036: guarded by a range of frames. Here is an example:
1037: .P1
1038: 1,5: pcp this %0
1039: pcp %0 that
1040: .P2
1041: This generates
1042: .P1
1043: pcp this %0
1044: pcp %0 that
1045: pcp this %0
1046: pcp %0 that
1047: pcp this %0
1048: pcp %0 that
1049: pcp this %0
1050: pcp %0 that
1051: pcp this %0
1052: pcp %0 that
1053: .P2
1054: The command group is repeated for each of frames 1 to 5.
1055: .PP
1056: Groups may contain parameter ranges enclosed in brackets
1057: .CW [] :
1058: .P1
1059: 1,5: pcp frame.[1,5] %0
1060: echo snap|2500
1061: .P2
1062: This generates:
1063: .P1
1064: pcp frame.1 %0
1065: echo snap|2500
1066: pcp frame.2 %0
1067: echo snap|2500
1068: pcp frame.3 %0
1069: echo snap|2500
1070: pcp frame.4 %0
1071: echo snap|2500
1072: pcp frame.5 %0
1073: echo snap|2500
1074: .P2
1075: .PP
1076: Programs may have multiple groups, each guarded by
1077: a separate range of frames. For each frame,
1078: .I moto
1079: checks each group and processes those
1080: whose guards include the current frame number.
1081: .PP
1082: Two special guards,
1083: .CW BEGIN
1084: and
1085: .CW END ,
1086: specify actions to be taken before an after processing frames:
1087: .P1
1088: BEGIN: clr
1089: 1,5: pcp section[1,5] %0
1090: END: pcp %0 composite
1091: .P2
1092: This generates
1093: .P1
1094: clr
1095: pcp section1 %0
1096: pcp section2 %0
1097: pcp section3 %0
1098: pcp section4 %0
1099: pcp section5 %0
1100: pcp %0 composite
1101: .P2
1102: .LP
1103: .I Moto
1104: allows complex computations inside parameter brackets:
1105: .P1 0
1106: 1,10: clr [127.5*(1-cos([0,360]))]
1107: .P2
1108: This generates
1109: .P1
1110: clr 0
1111: clr 29.82933350233
1112: clr 105.35985734747
1113: clr 191.25
1114: clr 247.3108091502
1115: clr 247.3108091502
1116: clr 191.25
1117: clr 105.35985734747
1118: clr 29.82933350233
1119: clr 0
1120: .P2
1121: .PP
1122: Expressions may include constants and variables.
1123: All values are double-precision floating point numbers.
1124: The operators
1125: .CW = ,
1126: .CW / ,
1127: .CW + ,
1128: .CW -
1129: (both unary and binary),
1130: .CW < ,
1131: .CW > ,
1132: .CW <= ,
1133: .CW >= ,
1134: .CW == ,
1135: .CW != ,
1136: .CW "? :"
1137: and
1138: .CW ! ,
1139: all with their meanings as in C, except that all results
1140: are coerced to
1141: .CW double .
1142: The result of
1143: .CW a%b
1144: is
1145: .CW a-b*(int)(a/b) .
1146: The result of
1147: .CW "a && b
1148: is
1149: .CW "a?b:a .
1150: The result of
1151: .CW "a || b
1152: is
1153: .CW "a?a:b .
1154: The exponentiation operator is
1155: .CW ^ ,
1156: also written
1157: .CW ** .
1158: The expression
1159: .CW [a,b]
1160: variesfrom
1161: .CW a
1162: to
1163: .CW b ,
1164: linearly as the frame number varies between the guards of the
1165: group containing the expression.
1166: The expression
1167: .CW a[b,c]
1168: has the value
1169: .CW a*b+(1-a)*c .
1170: Its value varies from
1171: .CW b
1172: to
1173: .CW c
1174: as
1175: .CW a
1176: varies from 0 to 1.
1177: The expression
1178: .CW $i
1179: has the value of the
1180: .CW i 'th
1181: parameter following the file name on
1182: .I moto 's
1183: command line.
1184: .PP
1185: The precedence of operators is, from lowest to highest:
1186: .P1
1187: =
1188: ? :
1189: ||
1190: &&
1191: < <= == != > >=
1192: + -
1193: * / %
1194: [ ]
1195: ^ **
1196: - \fR(unary)\fP ! $
1197: .P2
1198: Expressions may be parenthesized to alter precedence.
1199: .PP
1200: The following math functions are available:
1201: .QS
1202: .ft CW
1203: fabs floor ceil sqrt hypot sin cos tan
1204: asin acos atan exp log log10 sinh cosh tanh
1205: gamma besj0 besj1 besjn besy0 besy1 besyn
1206: .ft R
1207: .QE
1208: All math functions are as described in the C library,
1209: except that angles are measured in degrees rather than
1210: radians for the trig and inverse trig functions.
1211: In addition
1212: .CW hypot
1213: may have two or three arguments,
1214: .CW atan
1215: may take two arguments instead of one,
1216: and may also be spelled
1217: .CW atan2 .
1218: .PP
1219: For parameterization, and to allow even more complex
1220: computations,
1221: .I moto
1222: has variables, assignment and computation groups.
1223: A computation group is distinguished from a command group
1224: by having a double colon separating its guard from the
1225: expressions to be computed:
1226: .P1 0
1227: BEGIN:: n=5
1228: 1,n:: x=512*sin([0,90])
1229: 1,n: pcp -w 0 0 [x] 488 pic.[1,n] %0
1230: .P2
1231: This generates
1232: .P1 0
1233: pcp -w 0 0 0 488 pic.1 %0
1234: pcp -w 0 0 195.93391737093 488 pic.2 %0
1235: pcp -w 0 0 362.03867196751 488 pic.3 %0
1236: pcp -w 0 0 473.02632064578 488 pic.4 %0
1237: pcp -w 0 0 512 488 pic.5 %0
1238: .P2
1239: .PP
1240: Upon occasion it is useful to split
1241: .I moto 's
1242: output into several files, under program control.
1243: A group that is separated from its guards by an at-sign
1244: .CW @
1245: instead of a colon names a file into which
1246: subsequent output is to be written. For example,
1247: .P1
1248: 1,5@ file.[1,5]
1249: 1,5: This is file.[1,5].
1250: .P2
1251: creates 5 files, with names
1252: \fTfile.1\fR,...,\fTfile.5\fR.
1253: Each file's contents will announce its name.
1254: .PP
1255: As is true for all sufficiently large programs,
1256: .I moto
1257: has a shell escape. A group separated from its
1258: guards by an exclamation point
1259: .CW !
1260: instead of a colon has its result text interpreted
1261: by a subshell.
1262: .PP
1263: Finally, Figure 1 shows an example taken from a real application.
1264: This
1265: .I moto
1266: program composites the frames of a short movie showing
1267: two flying saucers, flying in formation, chased by a third,
1268: racing over New Jersey. The flying
1269: saucer images (files
1270: .CW run.*
1271: and
1272: .CW chase.* )
1273: and the background (file
1274: .CW bg )
1275: have been computed in advance. In the composite, the
1276: .CW run.*
1277: images are re-used, staggered in time, to do the
1278: first two saucers.
1279: .1C
1280: .KF
1281: .P1
1282: BEGIN:: nchase=108
1283: nrun=195
1284: d1=12
1285: d2=32
1286: end=nrun+d2
1287: chase=end-nchase+1
1288: 1,end: inputs= # empty the input list
1289: 1,nrun: inputs="$inputs run.[1,nrun]" # add the first saucer to the input list
1290: 1+d1,nrun+d1:
1291: inp="$inputs run.[1,nrun]" # add the second saucer
1292: chase,end:
1293: inp="$inputs chase.[1,nchase]" # add the chasing saucer
1294: 1,end:
1295: 3matte -a $inp bg frame.[1,end] # create the composite
1296: .P2
1297: .ce
1298: \fBFigure 1.\fP Flying saucer script
1299: .KE
1300: .2C
1301: .NH
1302: References
1303: .PP
1304: |reference_placement
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.