perm filename WHITNE.DOC[PIC,HE] blob sn#806012 filedate 1986-01-09 generic text, type T, neo UTF8
Whitney Image File Documentation:  September 6, 1984

	This file containts documentation for the images stored in
	[Whitney]/vis/pix.  Contained herein is a full directory
	listing, with README files inserted in appropriate places,
	and indenting used to make organization a little bit clearer.

	Send comments and suggestions to bradley@whitney.


/vis/pix/misc

This directory contains miscellaneous imagery ( some stereo ) from
miscellaneous sources.  All of the images should be in CMU format.

Images include:

	/vis/pix/misc/clrtst3.img
	/vis/pix/misc/clrtst3.img/blue
	/vis/pix/misc/clrtst3.img/green
	/vis/pix/misc/clrtst3.img/red

	/vis/pix/misc/tanks.img

	/vis/pix/misc/sf2.img

	/vis/pix/misc/washdc.r2.ster
	/vis/pix/misc/washdc.r2.ster/left
	/vis/pix/misc/washdc.r2.ster/right
	/vis/pix/misc/washdc.r2.ster/left.r2
	/vis/pix/misc/washdc.r2.ster/left.e2
	/vis/pix/misc/washdc.r2.ster/left.points
	/vis/pix/misc/washdc.r2.ster/left.r4
	/vis/pix/misc/washdc.r2.ster/left.r4.e8
	/vis/pix/misc/washdc.r2.ster/left.edges
	/vis/pix/misc/washdc.r2.ster/right.r2
	/vis/pix/misc/washdc.r2.ster/right.points
	/vis/pix/misc/washdc.r2.ster/right.r4
	/vis/pix/misc/washdc.r2.ster/right.r4.e8
	/vis/pix/misc/washdc.r2.ster/right.edges
	/vis/pix/misc/washdc.r2.ster/asterix.run.l

	/vis/pix/misc/portland.rgb
	/vis/pix/misc/portland.rgb/red
	/vis/pix/misc/portland.rgb/green
	/vis/pix/misc/portland.rgb/blue

	/vis/pix/misc/chair.rgb
	/vis/pix/misc/chair.rgb/red
	/vis/pix/misc/chair.rgb/green
	/vis/pix/misc/chair.rgb/blue

	/vis/pix/misc/sfobay.img

Stereo images include:
	/vis/pix/misc/ss2.stereo
	/vis/pix/misc/ss2.stereo/left
	/vis/pix/misc/ss2.stereo/right
	/vis/pix/misc/ss2.stereo/left.s
	/vis/pix/misc/ss2.stereo/right.s

	/vis/pix/misc/phone.stereo
	/vis/pix/misc/phone.stereo/left
	/vis/pix/misc/phone.stereo/right
	/vis/pix/misc/phone.stereo/left.r2
	/vis/pix/misc/phone.stereo/right.r2
	/vis/pix/misc/phone.stereo/left.points
	/vis/pix/misc/phone.stereo/left.edges
	/vis/pix/misc/phone.stereo/right.points
	/vis/pix/misc/phone.stereo/left.r4
	/vis/pix/misc/phone.stereo/left.r4.e8
	/vis/pix/misc/phone.stereo/right.edges
	/vis/pix/misc/phone.stereo/right.r4
	/vis/pix/misc/phone.stereo/right.r4.e8
	/vis/pix/misc/phone.stereo/left.e2
	/vis/pix/misc/phone.stereo/right.e2
	/vis/pix/misc/phone.stereo/asterix.run.l
	/vis/pix/misc/phone.stereo/asterix.ru.BAK

	/vis/pix/misc/sfquake.stereo
	/vis/pix/misc/sfquake.stereo/left
	/vis/pix/misc/sfquake.stereo/right
	/vis/pix/misc/sfquake.stereo/left.r2
	/vis/pix/misc/sfquake.stereo/right.r2
	/vis/pix/misc/sfquake.stereo/left.r4
	/vis/pix/misc/sfquake.stereo/right.r4

	/vis/pix/misc/washdc.stereo
	/vis/pix/misc/washdc.stereo/left
	/vis/pix/misc/washdc.stereo/right

	/vis/pix/misc/parts1.stereo
	/vis/pix/misc/parts1.stereo/left
	/vis/pix/misc/parts1.stereo/right
	/vis/pix/misc/parts2.stereo
	/vis/pix/misc/parts2.stereo/left
	/vis/pix/misc/parts2.stereo/right
	/vis/pix/misc/parts3.stereo
	/vis/pix/misc/parts3.stereo/left
	/vis/pix/misc/parts3.stereo/right
	/vis/pix/misc/parts4.stereo
	/vis/pix/misc/parts4.stereo/left
	/vis/pix/misc/parts4.stereo/right
	/vis/pix/misc/parts5.stereo


/vis/pix/etl

The files in this directory are sections of an image obtained from ETL on
tape.  Further information is available from Harlyn Baker.  The files were
created by Peter Blicher.

	/vis/pix/etl/ARIZ1A.RAW
	/vis/pix/etl/ARIZ1B.RAW
	/vis/pix/etl/ARIZ1C.RAW
	/vis/pix/etl/ARIZ1D.RAW
	/vis/pix/etl/ARIZ2A.RAW
	/vis/pix/etl/ARIZ2B.RAW
	/vis/pix/etl/ARIZ2C.RAW
	/vis/pix/etl/ARIZ2D.RAW

/vis/pix/ita
This directory contains all of the ITA imagery, both in the original
SRI image formats, and the CMU image format which we use.

The structure of the directory and its subdirectories are as follows:

	greyscale	-- this is a directory containing
			   all of the ITA greyscale images.
			   Each image in this directory
			   is in CMU image format.

	edge		-- this is a directory containing links to
			   all of the ITA edge greyscale images.
			   Each image in this directory
			   is in CMU image format.

	weldscan	-- this is a directory containing
			   all of the ITA weldscanner range images.
			   Each image in this directory is of
			   three components: x,y, and z.  They
			   are stored as three file named x, y, and
			   z in a directory bearing the image name.
			   Each component has been converted to
			   CMU image format.

	originals	-- this directory contains links to all
			   of the originals of all imagery shipped
			   over from SRI.  These images remain
			   in SRI image format.



	/vis/pix/ita/greyscale
	/vis/pix/ita/greyscale/poses1.img
	/vis/pix/ita/greyscale/poses2.img

	/vis/pix/ita/greyscale/itag1.img
	/vis/pix/ita/greyscale/itag2.img
	/vis/pix/ita/greyscale/itag3.img
	/vis/pix/ita/greyscale/ita5.img
	/vis/pix/ita/greyscale/ita6.img
	/vis/pix/ita/greyscale/ita7.img
	/vis/pix/ita/greyscale/ita8.img
	/vis/pix/ita/greyscale/ita5.edge
	/vis/pix/ita/greyscale/ita6.edge
	/vis/pix/ita/greyscale/ita7.edge
	/vis/pix/ita/greyscale/ita8.edge
	/vis/pix/ita/greyscale/ita8.ovr
	/vis/pix/ita/greyscale/ita5.vid
	/vis/pix/ita/greyscale/ita6.vid
	/vis/pix/ita/greyscale/ita7.vid
	/vis/pix/ita/greyscale/ita8.vid
	/vis/pix/ita/greyscale/ita5ed.vid
	/vis/pix/ita/greyscale/ita6ined.vid
	/vis/pix/ita/greyscale/ita6ed.vid
	/vis/pix/ita/greyscale/ita7ed.vid
	/vis/pix/ita/greyscale/ita8ed.vid
	/vis/pix/ita/greyscale/ita5ined.vid
	/vis/pix/ita/greyscale/ita7ined.vid
	/vis/pix/ita/greyscale/ita8ined.vid
	/vis/pix/ita/greyscale/itag1.inv
	/vis/pix/ita/greyscale/ita5.inv

	/vis/pix/ita/weldscan
		/vis/pix/ita/weldscan/ita.doc

	    Information about ITA range pictures
	    ------------------------------------
		As converted to CMU format
			--gangolli


	There are three pictures currently called ita1, ita3 and
	ita5.  For each picture there are three images (x, y and z),
	stored in files named x, y, and z in the subdirectories named
	ita1.img, ita3.img, ita5.img in this directory.

	All the images are coded as follows: 8bits/pixel with the
	convention:

		- missing data is coded 0 (zero)
		- regular data is coded from 50 to 255


	ita1    a 239 rows by 256 columns image.
	----

	Xmin = 75.6; Xmax = 81.7;   in the x image,
	Ymin = 20; Ymax = 22.1;     in the y image,
	Zmin = -19; Zmax = -17;     in the z image.

	ita3    a 273 rows by 256 columns image.
	----

	Xmin = 75.5; Xmax = 82.5;   in the x image,
	Ymin = 20; Ymax = 22.1;     in the y image,
	Zmin = -19; Zmax = -17;     in the z image.

	ita5    a 307 rows by 256 columns image.
	----

	Xmin = 75.5; Xmax = 83.3;   in the x image,
	Ymin = 20; Ymax = 22.1;     in the y image,
	Zmin = -19; Zmax = -17;     in the z image.

		/vis/pix/ita/weldscan/ita1.img
		/vis/pix/ita/weldscan/ita1.img/x
		/vis/pix/ita/weldscan/ita1.img/y
		/vis/pix/ita/weldscan/ita1.img/z
		/vis/pix/ita/weldscan/ita3.img
		/vis/pix/ita/weldscan/ita3.img/x
		/vis/pix/ita/weldscan/ita3.img/y
		/vis/pix/ita/weldscan/ita3.img/z
		/vis/pix/ita/weldscan/ita5.img
		/vis/pix/ita/weldscan/ita5.img/x
		/vis/pix/ita/weldscan/ita5.img/y
		/vis/pix/ita/weldscan/ita5.img/z

       /vis/pix/ita/originals 

       This directory contains all of the originals of images and
       processed data received from SRI for the ITA project.  Images
       are in SRI format.  The equivalents in CMU format reside in
       ../greyscale for greyscale images and ../weldscan for
       "weldscanner" data.  They were converted using the programs in
       /vis/local. (See /vis/local/doc/utilities.doc).

       There is a slight ambiguity in naming in this directory,
       because we chose to keep the names used by SRI.

       itag1.img to itag3.img, and poses?.img are the first set of
       greyscale photos received.  They have no accompanying
       extracted data.

       The directories ita5, ita6, ita7, and ita8 contain four more
       greyscale photos recieved from SRI. They are ./bw.img in each
       of these directories.  Also in these directories is stored
       extracted edge, arc, and chain information, as well as a
       readme.txt file which should be read.

       The images with names of the form ita?x.img, ita?y.img, and
       ita?z.img are the x,y, and z components of range information
       from SRI's "weldscanner".   They have NO relation to similarly
       named ita?.img greyscale images.

       You should also look at the other .doc files in this
       directory.

       srifmt.h is a C #include-able header file for the format of
       SRI images, and contains well-documented structure
       definitions.

		
	/vis/pix/ita/original/ita.doc

	    Information about ITA range pictures
	    ------------------------------------

       There are three pictures currently called ita1, ita3 and ita5.
       For each picture there are three images (x, y and z). The file
       names for these pictures are: (stored in this order on the
       tape)

	ita1:		ita1x.img
			ita1y.img
			ita1z.img
	ita3:		ita3x.img
			ita3y.img
			ita3z.img
	ita5:		ita5x.img
			ita5y.img
			ita5z.img

       All the images are coded as follows: 8bits/pixel with the
       convention:

		- missing data is coded 0 (zero)
		- regular data is coded from 50 to 255


       ita1    a 239 rows by 256 columns image.
	----

	Xmin = 75.6; Xmax = 81.7;   in the x image,
	Ymin = 20; Ymax = 22.1;     in the y image,
	Zmin = -19; Zmax = -17;     in the z image.

	ita3    a 273 rows by 256 columns image.
	----

	Xmin = 75.5; Xmax = 82.5;   in the x image,
	Ymin = 20; Ymax = 22.1;     in the y image,
	Zmin = -19; Zmax = -17;     in the z image.

	ita5    a 307 rows by 256 columns image.
	----

	Xmin = 75.5; Xmax = 83.3;   in the x image,
	Ymin = 20; Ymax = 22.1;     in the y image,
	Zmin = -19; Zmax = -17;     in the z image.

		/vis/pix/ita/originals/itag1.img
		/vis/pix/ita/originals/itag2.img
		/vis/pix/ita/originals/itag3.img

	/vis/pix/ita/originals/poses1.img
	/vis/pix/ita/originals/poses2.img

	Received: from SRI-AI.ARPA by SU-SCORE.ARPA with TCP; Tue 16
		  Aug 83 10:45:46-PDT
	Received: from SU-SCORE.ARPA by Whitney with TCP/SMTP; 16 Aug
		  83 10:45:54 PDT (Tue)
	Date: 16 Aug 1983 1040-PDT
	From: Hannah@SRI-AI
	Subject: Image format-1
	To: gangolli%whitney@SU-SCORE
	Cc: hannah@SU-SCORE.ARPA:HANNAH@SRI-AI.ARPA

	Anil--

	I am sending you (in a separate message: SEE srifmt.doc
	--gangolli) the information that we have on the testbed
	format.  It is all words, but I think you can pick the
	formula out of it.  (I'd send you the code, but it is in
	machine language, and there are several inter-related files,
	so it's messy.)

	The following lists the header field names and their contents
	for two of the files.  The fields are ASCII, 8-bytes wide;
	the '|' characters delimit the fields for the purposes of my
	explanation, only.  The beginning of the header is as
	follows: 


	field name	ita1z.img	itag1.img
	----------	---------	---------
	formattype	|SRI TB01|	|SRI TB01|
	recordtype	|IMGHDR  |	|IMGHDR  |
	recordbytes	|    1024|	|    1024|
	scantype	|LRBTSCAN|	|LRBTSCAN|
	bandtype	|UNSIGNED|	|UNSIGNED|
	imagebands	|       1|	|       1|
	bandbits	|       8|	|       8|
	pixelbits	|       8|	|       8|
	linepixels	|     256|	|     512|
	imagelines	|     239|	|     512|
	blockcols	|     256|	|      32|
	blockrows	|       1|	|      32|
	stripblocks	|       1|	|      16|
	imagestrips	|     239|	|      16|

	What this means is that the elevation image is in the most
	simple of formats-- it is a header, followed by 239 rows of
	256 pixels each.  The gray-scale image is blocked into 32x32
	chunks (32 rows of 32 pixels each).  The origin is at the
	lower left.  The only thing that might differ in later image
	deliveries is that the White Scanner produces 16-bit output;
	these images will probably be in the blocked format, so
	design your code accordingly.

	Good luck!
			mjh

	p.s.  You may want to clip the gray-scale images a bit.  In
	looking at them, I find that they were digitized over a test
	pattern, which included a GRINNELL logo at the top!
	-------


		/vis/pix/ita/originals/ita1x.img
		/vis/pix/ita/originals/ita1y.img
		/vis/pix/ita/originals/ita1z.img
		/vis/pix/ita/originals/ita3x.img
		/vis/pix/ita/originals/ita3y.img
		/vis/pix/ita/originals/ita3z.img
		/vis/pix/ita/originals/ita5x.img
		/vis/pix/ita/originals/ita5y.img
		/vis/pix/ita/originals/ita5z.img
		/vis/pix/ita/originals/srifmt.h


	/vis/pix/ita/originals/ita5

			Info for ita5:


	edges made:			11/29/83

	raw img:			.../pic/ita/gray/ita5/bw.img
	edge img:			.../pic/ita/gray/ita5/edge/bw.img
	edge magnitude img:		...no space on usr:
	linked chains:			.../pic/ita/gray/ita5/link/chains.lin

	std1:				2.0
	std2:				3.2
	zero crossing min magnitude:	0.5
	link←edges minchain:		10

	number of chains:		69
	number of edge points in chains:5333
	total number of edge points:	5616


	Camera calibration matrix:
	
	| u |	  |-77.1794	  20.626       -12.241 	   449.053 |  | x |
	| v |	= |-22.0919	 -82.7047       -6.0649    824.852 |  | y |
	| w |	  |   0		   0  	     	 0.931308     0    |  | z |
	| 1 |	  |  0.650539E-03 -0.857079E-03 -0.282579E-01 1	   |  | 1 |

	(x,y,z) = world coordinates 
	(u,v) = image coordinates

	Aspect Ratio:
		horizontal: 512 pixels = 6.36431 inches
		vertical: 512 pixels = 5.96714 inches

	NOTE: the data in these imgs has been hacked by hand to take
	out some of the junk around the boarders.

		/vis/pix/ita/originals/ita5/arcs.txt
		/vis/pix/ita/originals/ita5/arcslines.txt
		/vis/pix/ita/originals/ita5/bw.img
		/vis/pix/ita/originals/ita5/chains.txt
		/vis/pix/ita/originals/ita5/edge
		/vis/pix/ita/originals/ita5/edge/bw.img
		/vis/pix/ita/originals/ita5/edge/.emacs←118
		/vis/pix/ita/originals/ita5/lines.txt
		/vis/pix/ita/originals/ita5/link
		/vis/pix/ita/originals/ita5/link/chains.lin

	/vis/pix/ita/originals/ita6
			Info for ita6:


	edges made:			11/29/83

	raw img:			.../pic/ita/gray/ita6/bw.img
	edge img:			.../pic/ita/gray/ita6/edge/bw.img
	edge magnitude img:		...no space on usr:
	linked chains:			.../pic/ita/gray/ita6/link/chains.lin

	std1:				2.0
	std2:				3.2
	zero crossing min magnitude:	0.5
	link←edges minchain:		10

	number of chains:		76
	number of edge points in chains:5549
	total number of edge points:	5995

	Camera calibration matrix:
	
	| u |	  |-77.1794	  20.626       -12.241 	   449.053 |  | x |
	| v |	= |-22.0919	 -82.7047       -6.0649    824.852 |  | y |
	| w |	  |   0		   0  	     	 0.931308     0    |  | z |
	| 1 |	  |  0.650539E-03 -0.857079E-03 -0.282579E-01 1	   |  | 1 |

	(x,y,z) = world coordinates 
	(u,v) = image coordinates

	Aspect Ratio:
		horizontal: 512 pixels = 6.36431 inches
		vertical: 512 pixels = 5.96714 inches


	NOTE: the data in these imgs has been hacked by hand to take
	out some of the junk around the boarders.

		/vis/pix/ita/originals/ita6/arcs.txt
		/vis/pix/ita/originals/ita6/arcslines.txt
		/vis/pix/ita/originals/ita6/bw.img
		/vis/pix/ita/originals/ita6/chains.txt
		/vis/pix/ita/originals/ita6/edge
		/vis/pix/ita/originals/ita6/edge/bw.img
		/vis/pix/ita/originals/ita6/lines.txt
		/vis/pix/ita/originals/ita6/link
		/vis/pix/ita/originals/ita6/link/chains.lin

	/vis/pix/ita/originals/ita7
			Info for ita7:


	edges made:			11/29/83

	raw img:			.../pic/ita/gray/ita7/bw.img
	edge img:			.../pic/ita/gray/ita7/edge/bw.img
	edge magnitude img:		...no space on usr:
	linked chains:			.../pic/ita/gray/ita7/link/chains.lin

	std1:				4	2.0
	std2:				3.2
	zero crossing min magnitude:	0.5
	link←edges minchain:		10
	
	number of chains:		80
	number of edge points in chains:6660
	total number of edge points:	6986

	Camera calibration matrix:
	
	| u |	  |-77.1794	  20.626       -12.241 	   449.053 |  | x |
	| v |	= |-22.0919	 -82.7047       -6.0649    824.852 |  | y |
	| w |	  |   0		   0  	     	 0.931308     0    |  | z |
	| 1 |	  |  0.650539E-03 -0.857079E-03 -0.282579E-01 1	   |  | 1 |

	(x,y,z) = world coordinates 
	(u,v) = image coordinates

	Aspect Ratio:
		horizontal: 512 pixels = 6.36431 inches
		vertical: 512 pixels = 5.96714 inches


	NOTE: the data in these imgs has been hacked by hand to take
	out some of the junk around the boarders.

		/vis/pix/ita/originals/ita7/arcs.txt
		/vis/pix/ita/originals/ita7/arcslines.txt
		/vis/pix/ita/originals/ita7/bw.img
		/vis/pix/ita/originals/ita7/chains.txt
		/vis/pix/ita/originals/ita7/edge
		/vis/pix/ita/originals/ita7/edge/bw.img
		/vis/pix/ita/originals/ita7/lines.txt
		/vis/pix/ita/originals/ita7/link
		/vis/pix/ita/originals/ita7/link/chains.lin

	/vis/pix/ita/originals/ita8
			Info for ita8:


	edges made:			11/29/83

	raw img:			.../pic/ita/gray/ita8/bw.img
	edge img:			.../pic/ita/gray/ita8/edge/bw.img
	edge magnitude img:		...no space on usr:
	linked chains:			.../pic/ita/gray/ita8/link/chains.lin

	std1:				2.0
	std2:				3.2
	zero crossing min magnitude:	0.5
	link←edges minchain:		10

	number of chains:		82
	number of edge points in chains:5643
	total number of edge points:	5973

	Camera calibration matrix:
	
	| u |	  |-77.1794	  20.626       -12.241 	   449.053 |  | x |
	| v |	= |-22.0919	 -82.7047       -6.0649    824.852 |  | y |
	| w |	  |   0		   0  	     	 0.931308     0    |  | z |
	| 1 |	  |  0.650539E-03 -0.857079E-03 -0.282579E-01 1	   |  | 1 |

	(x,y,z) = world coordinates 
	(u,v) = image coordinates

	Aspect Ratio:
		horizontal: 512 pixels = 6.36431 inches
		vertical: 512 pixels = 5.96714 inches


	NOTE: the data in these imgs has been hacked by hand to take
	out some of the junk around the boarders.

	/vis/pix/ita/originals/ita8/arcs.txt
	/vis/pix/ita/originals/ita8/arcslines.txt
	/vis/pix/ita/originals/ita8/bw.img
	/vis/pix/ita/originals/ita8/chains.txt
	/vis/pix/ita/originals/ita8/edge
	/vis/pix/ita/originals/ita8/edge/bw.img
	/vis/pix/ita/originals/ita8/lines.txt
	/vis/pix/ita/originals/ita8/link
	/vis/pix/ita/originals/ita8/link/chains.lin

	/vis/pix/ita/Edge

	This directory is here for organizational purposes only it is
	linked to: /vis/pix/ita/greyscale since the images are
	greyscale images of edges.

	-- DMC 12/19/83

		/vis/pix/ita/edge/ita5.edge
		/vis/pix/ita/edge/ita6.edge
		/vis/pix/ita/edge/ita7.edge
		/vis/pix/ita/edge/ita8.edge
	
/vis/pix/leonie
	/vis/pix/leonie/partd.cmu
	/vis/pix/leonie/partd.r2.cmu
	/vis/pix/leonie/sandy.cmu
	
/vis/pix/leonie/cube.stereo
	/vis/pix/leonie/cube.stereo/left
	/vis/pix/leonie/cube.stereo/right
	/vis/pix/leonie/cube.stereo/left.r2
	/vis/pix/leonie/cube.stereo/left.points
	/vis/pix/leonie/cube.stereo/left.r4
	/vis/pix/leonie/cube.stereo/left.r4.e8
	/vis/pix/leonie/cube.stereo/left.edges
	/vis/pix/leonie/cube.stereo/right.r2
	/vis/pix/leonie/cube.stereo/right.points
	/vis/pix/leonie/cube.stereo/right.r4
	/vis/pix/leonie/cube.stereo/right.r4.e8
	/vis/pix/leonie/cube.stereo/right.edges
	/vis/pix/leonie/cube.stereo/left.e2
	/vis/pix/leonie/cube.stereo/right.e2
	/vis/pix/leonie/cube.stereo/asterix.run.l
	/vis/pix/leonie/demo.cmu
	
/vis/pix/roads
	Color images for road following

	naming xyn.rgb
	  x l = left image, r = right image
	  y series: 
	    b : around D.C. Power Lab
	    c : right side(south) of slac
	    d : left side (north) of slac
	  n step in  sequence, one step is about 3 meters

	use shell script 
		shape image.rgb
	(in this file: calls shapeup)
	to creat CMU color image triple
	display with dispc
	put on tape with tar

	/vis/pix/roads/lc1.rgb
		/vis/pix/roads/lc1.rgb/b
		/vis/pix/roads/lc1.rgb/g
		/vis/pix/roads/lc1.rgb/r
		/vis/pix/roads/lc1.rgb/red
		/vis/pix/roads/lc1.rgb/green
		/vis/pix/roads/lc1.rgb/blue
	/vis/pix/roads/lc2.rgb
		/vis/pix/roads/lc2.rgb/r
		/vis/pix/roads/lc2.rgb/g
		/vis/pix/roads/lc2.rgb/b
		/vis/pix/roads/lc2.rgb/red
		/vis/pix/roads/lc2.rgb/green
		/vis/pix/roads/lc2.rgb/blue
	/vis/pix/roads/ld3.rgb
		/vis/pix/roads/ld3.rgb/blue
		/vis/pix/roads/ld3.rgb/green
		/vis/pix/roads/ld3.rgb/red
	/vis/pix/roads/ld4.rgb
		/vis/pix/roads/ld4.rgb/red
		/vis/pix/roads/ld4.rgb/green
		/vis/pix/roads/ld4.rgb/blue
	/vis/pix/roads/rc1.rgb
		/vis/pix/roads/rc1.rgb/green
		/vis/pix/roads/rc1.rgb/blue
		/vis/pix/roads/rc1.rgb/red
	/vis/pix/roads/rc2.rgb
		/vis/pix/roads/rc2.rgb/red
		/vis/pix/roads/rc2.rgb/green
		/vis/pix/roads/rc2.rgb/blue
	/vis/pix/roads/rd3.rgb
		/vis/pix/roads/rd3.rgb/red
		/vis/pix/roads/rd3.rgb/green
		/vis/pix/roads/rd3.rgb/blue
	/vis/pix/roads/rd4.rgb
		/vis/pix/roads/rd4.rgb/red
		/vis/pix/roads/rd4.rgb/green
		/vis/pix/roads/rd4.rgb/blue
	/vis/pix/roads/rb4.rgb
		/vis/pix/roads/rb4.rgb/b
		/vis/pix/roads/rb4.rgb/g
		/vis/pix/roads/rb4.rgb/r
		/vis/pix/roads/rb4.rgb/red
		/vis/pix/roads/rb4.rgb/green
		/vis/pix/roads/rb4.rgb/blue
	/vis/pix/roads/lb4.rgb
		/vis/pix/roads/lb4.rgb/red
		/vis/pix/roads/lb4.rgb/green
		/vis/pix/roads/lb4.rgb/blue
	/vis/pix/roads/shape
	/vis/pix/roads/rb5.rgb
		/vis/pix/roads/rb5.rgb/b
		/vis/pix/roads/rb5.rgb/g
		/vis/pix/roads/rb5.rgb/r
	/vis/pix/roads/lb5.rgb
		/vis/pix/roads/lb5.rgb/red
		/vis/pix/roads/lb5.rgb/green
		/vis/pix/roads/lb5.rgb/blue
	/vis/pix/roads/rb6.rgb
		/vis/pix/roads/rb6.rgb/b
		/vis/pix/roads/rb6.rgb/g
		/vis/pix/roads/rb6.rgb/r
	/vis/pix/roads/lb6.rgb
		/vis/pix/roads/lb6.rgb/red
		/vis/pix/roads/lb6.rgb/green
		/vis/pix/roads/lb6.rgb/blue
	
/vis/pix/ernst
	contains images used by ernst

	parts: various assembly line parts
	partb:    "
	partc:   " 13 parts, 100*100 grey min= 58, mean 154 max 255,
		meandif 10
	sandy: lab with sandys wive
	tanks; oil tanks 128*128 grey levels 90 .. 220
	g, p, z: temporary copies

		/vis/pix/ernst/tanks
		/vis/pix/ernst/sandy
		/vis/pix/ernst/parts
		/vis/pix/ernst/partd
		/vis/pix/ernst/partc
		/vis/pix/ernst/partb
		/vis/pix/ernst/p
		/vis/pix/ernst/cube64
		/vis/pix/ernst/mountain

	/vis/pix/ernst/phone

	The stereo phone picture l(eft) & r(right) reduced by averaging 
	e.g. l2 = left image reduced by a factor of 2 (128 * 128)
	     l4 = 64*64  l8 = 32*32 r16 = 16*16
	grey level statistics: max = 200, mean = 100, min = 0
			       (13% stat. black phone)
	max neighbor difference = 140 largely constant over resolution
	Taking l2 as reference:
	 Tiled background wall:
	  grey levels of tiles = 130 ... 200
	  with of vertical stripes = 2 pixel
	  disparity (l2 - r2) = -19 (or 2, or 22)
	  position of upper row of stripes in l2 (greylevels)
	   5(18-40) 18(~150) 26(50-60) 46(80-90)
					   in r2
	  24(40-50)    (125) 48(80-90) 70
	  in the upper left corner r2 is brighter by about 25 and up to 60 than
	  corresponding parts of l2.
	Phone keys: the upper left key is at (69,73) in l2 and (68,72) in r2
	 brightest points of keys are 178 - 182 in l2
	  the same range for r2 but individual differences of +- 2
	 darkest spots between keys 65 to 75 in both images
	 Distance between keys: 5 cols, 4 rows

		/vis/pix/ernst/phone/l1
		/vis/pix/ernst/phone/r1
		/vis/pix/ernst/phone/l2
		/vis/pix/ernst/phone/r2
		/vis/pix/ernst/phone/r4
		/vis/pix/ernst/phone/r8
		/vis/pix/ernst/phone/r16
		/vis/pix/ernst/phone/l4
		/vis/pix/ernst/phone/l8
		/vis/pix/ernst/phone/l4.e8
		/vis/pix/ernst/phone/r4.e8
		/vis/pix/ernst/phone/l1.e8
		/vis/pix/ernst/phone/l2.e8
		/vis/pix/ernst/phone/r2.e8
		/vis/pix/ernst/phone/z
		/vis/pix/ernst/phone/z1
		/vis/pix/ernst/phone/z2
		/vis/pix/ernst/phone/l4.e8c
		/vis/pix/ernst/phone/r4.e8c

	/vis/pix/ernst/cube64.e8

	/vis/pix/ernst/partc.e8
	/vis/pix/ernst/cube512
	/vis/pix/ernst/cube128
	/vis/pix/ernst/cube128.e8
	/vis/pix/ernst/square32.e8
	/vis/pix/ernst/square
	/vis/pix/ernst/square128
	/vis/pix/ernst/square64
	/vis/pix/ernst/square32
	/vis/pix/ernst/cube32
	/vis/pix/ernst/cube32.e8
	/vis/pix/ernst/noise256
	/vis/pix/ernst/tanks.e8
	/vis/pix/ernst/edges.l
	/vis/pix/ernst/chair64
	
	/vis/pix/vic
	/vis/pix/vic/SF
	/vis/pix/vic/tools
	/vis/pix/vic/toolsM
	/vis/pix/vic/toolsS
	/vis/pix/vic/toolsE
	/vis/pix/vic/phoneEC
	/vis/pix/vic/SFE
	/vis/pix/vic/SFS
	/vis/pix/vic/phone
	/vis/pix/vic/phoneE
	/vis/pix/vic/phoneM
	/vis/pix/vic/phoneS