perm filename PIX.DOC[PIC,HE]3 blob
sn#367358 filedate 1978-07-13 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00016 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00003 00002 Picture Data Base
C00004 00003 B&W 9x9 aerial photos, Sacremento sewer survey.
C00011 00004 Color 9x9 aerial San Mateo housing development.
C00013 00005 TANK AND TRUCK PIX, monocular
C00018 00006 SRI PIX, monocular
C00019 00007 CMU Pittsburgh scenes. Stereo and Color
C00020 00008 CDC Block Form, Synthesized City Scene
C00021 00009 B&W 9x9 aerial high altitude SF airport.
C00022 00010 San Francisco Airport Stereo, Black and White
C00025 00011 CDC Series Power Plant Many More available
C00026 00012 LOCKHEED, SAR and visible
C00027 00013 People monocular, black and white
C00028 00014 Hughes Registered Multi-Sensor images
C00031 00015 Lockheed picture format
C00036 00016 correspondence
C00077 ENDMK
C⊗;
Picture Data Base
B&W 9x9 aerial photos, Sacremento sewer survey.
Parking Lot and Buildings, City in stereo
file format edg? noise
comment stereo pair aerial view of parking lot;
* SP2.USC SP2.SU 5.00+0; 3min Right view.
512X512x8
Hueckel requires 3 min, Walsh requires 4.5 sec (2.7 sec w/o i/o)
* SP3.USC SP3.SU 4.3+0; 3m20s Left view.
512X512x8
stereo pair aerial view of building with parking lot;
* S1P2.USC S1P2.SU (negative) 7.86
1024x512
* S1P3.USC S1P2.SU (negative); 8.57
1024x512
stereo pair aerial view of city scene. (large scale)
A1.USC A1.SU 8.57+0noise Left view.
512X512x8
A2.USC A2.SU 18.57+0noise Right view.
512X512x8
SB1.USC negative version of A1.USC
SB2.USC negative version of A2.USC
SP3 USC PIC HE 64.0 23-JUL-76 1506 000 1TOB COPY 23-JUL-76
SP2 USC PIC HE 64.0 23-JUL-76 1507 000 1TOB COPY 23-JUL-76
S1P2 USC PIC HE 256.0 23-JUL-76 1508 000 1TOB COPY 23-JUL-76
S1P3 USC PIC HE 256.0 23-JUL-76 1513 000 1TOB COPY 23-JUL-76
SB1 USC PIC HE 64.0 23-JUL-76 1517 000 1TOB COPY 23-JUL-76
SB2 USC PIC HE 64.0 23-JUL-76 1518 000 1TOB COPY 23-JUL-76
SP2 SU PIC HE 64.1 30-JUL-76 1531 400 1TOB COPY 13-DEC-76
SP3 SU PIC HE 64.1 30-JUL-76 1532 400 1TOB COPY 13-DEC-76
S1P2 SU PIC HE 128.1 30-JUL-76 1530 400 1TOB COPY 16-MAR-77
S1P3 SU PIC HE 128.1 30-JUL-76 1533 400 1TOB COPY 16-MAR-77
SB1 SU PIC HE 64.1 30-JUL-76 1524 400 1TOB COPY 13-DEC-76
SB2 SU PIC HE 64.1 30-JUL-76 1526 400 1TOB COPY 13-DEC-76
A1 SU PIC HE 64.1 23-JUL-76 1540 000 1TOB COPY 13-DEC-76
A2 SU PIC HE 64.1 23-JUL-76 1539 000 1TOB COPY 13-DEC-76
A1 HEF PIC HE 43.2 13-DEC-76 0932 000 1RDA COPY 13-DEC-76
A2 HEF PIC HE 43.2 13-DEC-76 0932 000 1RDA COPY 13-DEC-76
To be deleted.
A3A USC PIC HE 64.0 23-JUL-76 1519 000 1TOB COPY 23-JUL-76
A3B USC PIC HE 64.0 23-JUL-76 1530 000 1TOB COPY 23-JUL-76
S1 USC PIC HE 128.0 23-JUL-76 1502 000 1TOB COPY 23-JUL-76
A3A SU PIC HE 64.1 30-JUL-76 1523 400 1TOB COPY 30-JUL-76
A3B SU PIC HE 64.1 30-JUL-76 1528 400 1TOB COPY 30-JUL-76
S1 SU PIC HE 64.1 23-JUL-76 1541 000 1TOB COPY 30-JUL-76
Color 9x9 aerial San Mateo housing development.
comment these three files are a color triple
BCRED.USC BCRED.SU
1024x512
BCGRN.USC BCGRN.SU
BCBLU.USC BCBLU.SU
comment the following three files are a color triple
AERED.USC AERED.SU
512x512
AEGRN.USC AEGRN.SU
AEBLU.USC AEBLU.SU
AERED USC PIC HE 64.0 23-JUL-76 1520 000 1TOB COPY 23-JUL-76
AEGRN USC PIC HE 64.0 23-JUL-76 1521 000 1TOB COPY 23-JUL-76
AEBLU USC PIC HE 64.0 23-JUL-76 1522 000 1TOB COPY 23-JUL-76
BCRED USC PIC HE 256.0 23-JUL-76 1523 000 1TOB COPY 23-JUL-76
BCGRN USC PIC HE 256.0 23-JUL-76 1531 000 1TOB COPY 23-JUL-76
BCBLU USC PIC HE 256.0 23-JUL-76 1534 000 1TOB COPY 23-JUL-76
AERED SU PIC HE 64.1 23-JUL-76 1555 000 1TOB COPY 23-JUL-76
AEGRN SU PIC HE 64.1 23-JUL-76 1556 000 1TOB COPY 23-JUL-76
AEBLU SU PIC HE 64.1 23-JUL-76 1557 000 1TOB COPY 23-JUL-76
BCRED SU PIC HE 128.1 23-JUL-76 1557 000 1TOB COPY 23-JUL-76
BCGRN SU PIC HE 128.1 23-JUL-76 1601 000 1TOB COPY 23-JUL-76
BCBLU SU PIC HE 128.1 23-JUL-76 1603 000 1TOB COPY 23-JUL-76
TANK AND TRUCK PIX, monocular
* means we have a hard copy picture
<JOHNSTON>TANK.*.*
*<JOHNSTON>TANK.POS1;1 TANK11.USC TANK11.SU 4.29;
subject truck in desert
*<JOHNSTON>TANK.POS1;2 TANK12.USC TANK11.SU 2.14
subject plane parked in desert
<JOHNSTON>TANK.POS1;3 TANK13.USC
tank front view
<JOHNSTON>TANK.POS1;4 TANK14.USC
tank overhead
<JOHNSTON>TANK.POS2;1 TANK21.USC
3 trucks in a field
<JOHNSTON>TANK.POS2;2 TANK22.USC
3 trucks from different view
<JOHNSTON>TANK.POS2;3 TANK23.USC
3 vehicles
*<JOHNSTON>TANK.POS2;4 TANK24.USC
3 vehicles
*<JOHNSTON>TANK.POS3;1 TANK31.USC 7.14;
tank side view
*<JOHNSTON>TANK.POS3;2 TANK32.USC
military vehicle side view
<JOHNSTON>KLEIN.PC1;1 KLNC1.USC
<JOHNSTON>KLEIN.PC2;1 KLNC2.USC
<JOHNSTON>KLEIN.PC5;1 KLNC5.USC
<JOHNSTON>KLEIN.PB7;1 KLNB7.USC
<JOHNSTON>KLEIN.PB8;1 KLNB8.USC
<JOHNSTON>KLEIN.PD7;1 KLND7.USC
TANK11 USC PIC HE 64.0 30-JUL-76 1452 400 1TOB COPY 30-JUL-76
TANK12 USC PIC HE 64.0 30-JUL-76 1453 400 1TOB COPY 30-JUL-76
TANK13 USC PIC HE 64.0 30-JUL-76 1455 400 1TOB COPY 30-JUL-76
TANK14 USC PIC HE 64.0 30-JUL-76 1456 400 1TOB COPY 30-JUL-76
TANK24 USC PIC HE 64.0 30-JUL-76 1459 400 1TOB COPY 30-JUL-76
TANK23 USC PIC HE 64.0 30-JUL-76 1501 400 1TOB COPY 30-JUL-76
TANK22 USC PIC HE 64.0 30-JUL-76 1503 400 1TOB COPY 30-JUL-76
TANK21 USC PIC HE 64.0 30-JUL-76 1505 400 1TOB COPY 30-JUL-76
TANK31 USC PIC HE 64.0 30-JUL-76 1507 400 1TOB COPY 30-JUL-76
TANK32 USC PIC HE 64.0 30-JUL-76 1509 400 1TOB COPY 30-JUL-76
KLNC1 USC PIC HE 64.0 30-JUL-76 1511 400 1TOB COPY 30-JUL-76
KLNC2 USC PIC HE 64.0 30-JUL-76 1514 400 1TOB COPY 30-JUL-76
KLNC5 USC PIC HE 64.0 30-JUL-76 1516 400 1TOB COPY 30-JUL-76
KLNB8 USC PIC HE 64.0 30-JUL-76 1517 400 1TOB COPY 30-JUL-76
KLNB7 USC PIC HE 64.0 30-JUL-76 1518 400 1TOB COPY 30-JUL-76
KLND7 USC PIC HE 64.0 30-JUL-76 1520 400 1TOB COPY 30-JUL-76
TANK11 SU PIC HE 64.1 30-JUL-76 1535 400 1TOB COPY 13-DEC-76
TANK12 SU PIC HE 64.1 30-JUL-76 1536 400 1TOB COPY 13-DEC-76
TANK13 SU PIC HE 64.1 30-JUL-76 1537 400 1TOB COPY 13-DEC-76
TANK14 SU PIC HE 64.1 30-JUL-76 1537 400 1TOB COPY 13-DEC-76
TANK21 SU PIC HE 64.1 30-JUL-76 1538 400 1TOB COPY 13-DEC-76
TANK22 SU PIC HE 64.1 30-JUL-76 1539 400 1TOB COPY 13-DEC-76
TANK23 SU PIC HE 64.1 30-JUL-76 1558 400 1TOB COPY 13-DEC-76
TANK24 SU PIC HE 64.1 30-JUL-76 1601 400 1TOB COPY 13-DEC-76
TANK31 SU PIC HE 64.1 30-JUL-76 1603 400 1TOB COPY 30-JUL-76
TANK32 SU PIC HE 64.1 30-JUL-76 1604 400 1TOB COPY 30-JUL-76
KLNC1 SU PIC HE 64.1 30-JUL-76 1605 400 1TOB COPY 30-JUL-76
KLNC2 SU PIC HE 64.1 30-JUL-76 1606 400 1TOB COPY 30-JUL-76
KLNC5 SU PIC HE 64.1 30-JUL-76 1609 400 1TOB COPY 30-JUL-76
KLNB7 SU PIC HE 64.1 30-JUL-76 1610 400 1TOB COPY 30-JUL-76
KLNB8 SU PIC HE 64.0 30-JUL-76 1611 400 1TOB COPY 30-JUL-76
KLND7 SU PIC HE 64.1 30-JUL-76 1614 400 1TOB COPY 30-JUL-76
SRI PIX, monocular
on cart (BGB&TED) udp, under [AER,HPM], as SRIn.ij, 0≤i,j≤3
CMU Pittsburgh scenes. Stereo and Color
CMU12R.DAT RED BAND
CMU12G.DAT GREEN BAND
CMU12B.DAT BLUE BAND
CMU12I.DAT INTENSITY BAND (COMPOSITE OF RGB)
CMU15R.DAT RED BAND
CMU15G.DAT GREEN BAND
CMU15B.DAT BLUE BAND
CMU15I.DAT INTENSITY BAND (COMPOSITE OR RGB)
CMU12R CMU PIC HE 119.7 05-AUG-77 1731 000 1TOB COPY 05-AUG-77
CMU12G CMU PIC HE 119.7 05-AUG-77 1731 000 1TOB COPY 05-AUG-77
CMU12B CMU PIC HE 119.7 05-AUG-77 1731 400 1TOB COPY 05-AUG-77
CMU15G CMU PIC HE 119.7 05-AUG-77 1732 400 1TOB COPY 05-AUG-77
CMU15B CMU PIC HE 119.7 05-AUG-77 1732 400 1TOB COPY 05-AUG-77
CDC Block Form, Synthesized City Scene
CDC1 PIC HE 43.4 19-MAY-77 1639 400 1RDA COPY 19-MAY-77
CDC2 PIC HE 43.4 19-MAY-77 1641 400 1RDA COPY 19-MAY-77
CDC1 SU PIC HE 43.1 04-OCT-77 1252 400 1RDA COPY 04-OCT-77
CDC2 SU PIC HE 43.1 04-OCT-77 1253 400 1RDA COPY 04-OCT-77
B&W 9x9 aerial high altitude SF airport.
Airport terminal.
AI1.PIC AI1.SU
512x512x9 top view, some lines dropped out.
AI2.PIC AI2.SU
512x512x9 bottom view, some lines dropped out.
AI1 PIC PIC HE 129.5 28-APR-77 1401 400 1RDA COPY 28-APR-77
AI2 PIC PIC HE 129.5 28-APR-77 1402 400 1RDA COPY 28-APR-77
AI1 SU PIC HE 64.1 28-APR-77 1404 400 1RDA COPY 28-APR-77
AI2 SU PIC HE 64.1 28-APR-77 1404 400 1RDA COPY 28-APR-77
San Francisco Airport Stereo, Black and White
SF1 and SF4 are a pair of large scale pictures of the terminal; SF2 and
SF5 are a pair of small scale pictures of a single plane (L-1011, SF5 is right hand
view); SF3 is a small scale picture of a single plane (727?).
file format edg? noise
LJ.SU 128x128 2.86
this file reduced from SF2.SU[PIC,HE]
RJ.SU 128x128
this file reduced from SF5.SU[PIC,HE]
These pictures are 512x512x9 unpacked. High order bit seems to be always 0.
FILE1.PIC, FILE2.PIC, FILE3.PIC, FILE4.PIC, FILE5.PIC
These pictures converted from FILE1.PIC, FILE2.PIC, etc. All are 512x512x8.
SF2.SU 1.43 st dev
SF5.SU 1.43 st dev
SF1.SU left view of terminal
SF2.SU left view of single plane, L-1011
SF3.SU left view of 1 1/2 planes
SF4.SU right view of terminal
SF5.SU right view of single plane, L-1011
FILE1 PIC PIC HE 129.5 04-OCT-77 1243 400 1RDA COPY 04-OCT-77
FILE2 PIC PIC HE 129.5 04-OCT-77 1245 400 1RDA COPY 04-OCT-77
FILE3 PIC PIC HE 129.5 04-OCT-77 1246 400 1RDA COPY 04-OCT-77
FILE4 PIC PIC HE 129.5 04-OCT-77 1247 400 1RDA COPY 04-OCT-77
FILE5 PIC PIC HE 129.5 04-OCT-77 1248 400 1RDA COPY 04-OCT-77
SF1 SU PIC HE 64.1 04-OCT-77 1254 400 1RDA COPY 04-OCT-77
SF2 SU PIC HE 64.1 04-OCT-77 1255 400 1RDA COPY 04-OCT-77
SF3 SU PIC HE 64.1 04-OCT-77 1256 400 1RDA COPY 04-OCT-77
SF4 SU PIC HE 64.1 04-OCT-77 1256 400 1RDA COPY 04-OCT-77
SF5 SU PIC HE 64.1 04-OCT-77 1256 400 1RDA COPY 04-OCT-77
CDC Series Power Plant; Many More available
CDCA HEF PIC HE 4.3 13-DEC-76 0932 000 1RDA COPY 13-DEC-76
CDCB HEF PIC HE 4.3 13-DEC-76 0932 000 1RDA COPY 13-DEC-76
CDCC HEF PIC HE 4.3 13-DEC-76 0932 000 1RDA COPY 13-DEC-76
CDCD HEF PIC HE 4.3 13-DEC-76 0932 000 1RDA COPY 13-DEC-76
Two views of power plant, at right angles to each other, oblique, one A2 very noisy.
A1.HEF
A2.HEF
A1 HEF PIC HE 43.2 13-DEC-76 0932 000 1RDA COPY 06-OCT-77
A2 HEF PIC HE 43.2 13-DEC-76 0932 000 1RDA COPY 06-OCT-77
LOCKHEED, SAR and visible
The Photo and SAR pictures from Lockheed now are on the SYS UDP as
PHOTO.n[PIC,HE] and SAR.n[PIC,HE], where n = 1 to 6.
People monocular, black and white
JOMARY.SU JOMARY.EDG 1.0+0noise
287x216 47 sec for edges;
subject face of a woman; 6 bit TV
JOMARY SU PIC HE 10.2 23-JUL-76 1540 000 1TOB COPY 30-JUL-76
Hughes Registered Multi-Sensor images
I1, I2, I3, I4, I5 image names
V=visible, R=infrared, 3=3mm active, 10=10mm active
I2BV HEF PIC HE 9.7 11-FEB-77 1055 000 1TOB COPY 11-FEB-77
I2BR HEF PIC HE 9.7 11-FEB-77 1055 000 1TOB COPY 11-FEB-77
I2B10 HEF PIC HE 9.7 11-FEB-77 1055 000 1TOB COPY 11-FEB-77
I2B3 HEF PIC HE 9.7 11-FEB-77 1055 000 1TOB COPY 11-FEB-77
I1BV HEF PIC HE 14.0 11-FEB-77 1055 000 1TOB COPY 16-MAR-77
I1BR HEF PIC HE 7.0 11-FEB-77 1055 000 1TOB COPY 11-FEB-77
I1B10 HEF PIC HE 7.p 11-FEB-77 1055 000 1TOB COPY 11-FEB-77
I1B3 HEF PIC HE 7.0 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I3BV HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I3BR HEF PIC HE 17.5 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I3B10 HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I3B3 HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I4BV HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I4BR HEF PIC HE 17.5 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I4B10 HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I4B3 HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I5BV HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I5BR HEF PIC HE 17.5 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I5B10 HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
I5B3 HEF PIC HE 8.7 11-FEB-77 1056 000 1TOB COPY 11-FEB-77
L015-F HEF PIC HE 9.7 11-FEB-77 1055 000 1TOB COPY 19-MAY-77
Lockheed picture format
Tape format
2052 words/record integers
16 bits/word
for 512x512 pixel images
they have 4 lines/record (with 4 unused words)
lines are horizontal
9 bits per pixel in range 0-511
Text of header file in words 31-70
Image file header record info
word name meaning
1 ipsb storage bits per pixel
2 isize code for image size quanta
3 ixl horizontal file length in pixels
4 iyl vertical file length in pixels
5 imode file storage mode (horiz or vertical)
6 ipib information bits per pixel
7-9 date month, day,, and year from rdos system call
10-12 time hour, minute, and second from rdos system call
13 ipps pixel pre-pack bit shift amount
14 ipus pixel post-unpack bit shift amount
15 ixs horizontal starting position
16 iys vertical starting position
17 idt data type of pixel elements
18 isf scale factor of pixel elements
19 ipob offset bits (low-order) per pixel
20 imsk pixel pre-display post-shift "and" mask
21 ihvn header format version number
22 nwe number of words/element
23-30 currently undefined; set to zero
31-70 text text string for user comments
71-256 currently undefined; set to zero
Explanations
Header Format Evolution: The header format
version number defines what the rest of the header rcord
contains and how it is to be interpreted.
For version 1: Th pixel data format is defined by ipsb, ipib, ipob, idt, and imsk.
The stored pixel values are not packed, ie, ther is one pixel per word
always. Within the word, the pixel value can or will be shifted
right or left when displaying and/or keeping. The subroutine ALTER
is used for shifting and/or masking; if neither operation is wanted,
the data is ready in the same buffer and no time is wasted moving and preparing
data. For displaying, the "and" mask IMSK is applied--after any shifting
is finished to remove undesired bits in the word.
An imsk value of 0 causes the masking operation to be avoided; it is logically
equivalent to a value of -1.
Data model ----xxxxxxxx----
ipib=8 ipob=4
ipsb=16
For any data movement operation, a default signed shift is automatically
computed and undertaken unless the user explicitly specifies an alternate
value to be used. The default shift amount--given by the equation
below--is designed so that alignment occurs with respect to the highest-order
information bit. Scanner core pixel data format is assumed to be ipsb=16,
ipib=9, and ipob=8. ishift=(ipib+ipob)new-(ipib+ipob)old.
Note that ipps and ipus are not utilized for header format version 1.
correspondence
FILES ARE VISJ:CMU12?.DAT[A640DB0I] AND VISJ:CMU15?.DAT[A640DB0I]
MCKEOWN%CMU-10B
TOM
RAJ REDDY HAS ASKED ME TO GIVE YOU A POINTER TO
TWO DIGITIZED PICTURES OF DOWNTOWN PITTSBURGH SUITABLE
FOR GENERATION OF A DEPTH IMAGE.
THEY ARE IN THE FOLLOWING FORMAT:
CMU12R.DAT RED BAND
CMU12G.DAT GREEN BAND
CMU12B.DAT BLUE BAND
CMU12I.DAT INTENSITY BAND (COMPOSITE OF RGB)
CMU15R.DAT RED BAND
CMU15G.DAT GREEN BAND
CMU15B.DAT BLUE BAND
CMU15I.DAT INTENSITY BAND (COMPOSITE OR RGB)
EACH FILE IS COMPOSED OF A 128 WORD HEADER (WHICH
YOU CAN IGNORE) FOLLOWED BY A 700 X 700 PIXEL IMAGE
ALL PIXELS ARE 8 BITS EXCEPT FOR THE INTENSITY
IMAGE WHICH HAS BEEN NORMALIZED TO 4 BITS FOR OUR
GOULD PRINTER. THE PIXELS ARE PACKED USING
NORMAL PDP10 BYTE CONVENTIONS.
EACH IS ABOUT 970 BLOCKS (EXCEPT FOR THE
INTENSITY FILES WHICH ARE ABOUT HALF AS LARGE).
UNFORTUNITLY THEY RESIDE ON A PRIVATE PACK WHICH
IS NOT NORMALLY MOUNTED. LET ME KNOW
WHEN YOU WOULD LIKE TO GET THE FILES (VIA FTP)
AND I WILL HAVE THE PACK MOUNTED AND WILL BE
AROUND IN CASE THERE ARE PROBLEMS.
SEND MAIL TO MCKEOWN@CMUB
HOPE TO HEAR FROM YOU SOON
DAVE
-------
pix
I have transferred the tape program to TAPE.SAI[1,PDQ] for your reference.
I have transferred the file to AI Lab file AI1.PIC[1,PDQ].
NEVATIA
It is difficult to predict a turn around time, but I can try
to have your pictures digitized on a priority basis. How soon
do you need them? If pictures from prints are acceptable, a
much faster job can be done on the drum scanner.
We are in a short supply of good vision students here. Do you
have any that you are unable to support who would consider a
transfer (is there a reject list available?)?
Ram.
-------
NCIC National Cartographic Info Center
Ames Souix Falls Center U2 footage
Data Facility Bob Ekstrom, Jim Ryan 965-6252
Earth Resources Aircraft Proj Office
they have 270,000 frames of U2 from 65,000 ft
in the form of transparencies
LANDSAT ARCHIVES OF BAY AREA. PROF RON LYONS, STANFORD
∂18-FEB-76 1342 RDA
Custom Color Laboratory will make an 8x8 transparency for $15.
Color Tech Corp. will make a 4x5 negative for $7 and an 8x10
transparency for $25. (In which case they make the 4x5 neg anyway.)
I cannot find a reasonable xxx for "International xxx" with a
Sunnyvale address.
Western Aerial Photo
Redwood City 369-5273
gave some free pix
Dean Powell, Sacramento Sewer Survey
303 Whipple Ave.
Air Photo Co., Inc
2632 Bayshore Frontage Rd.
M. View, 969-2671
no freebies
Gene
Black and white $135 for flight + $16 per exposure
Color $275 for flight + $16 per exposure
Resolution
6" and 12" focal length Wild RC10 9"x9"
100 lines
1200 feet minimum altitude; they have gotten clearance over SF airport
they have mapped it several times
Aerometrics 56-347-7009
Mr. Rankle still in progress
2100 ft with 6" focal length
freight loading areas probably best bet
Mr. Dick Foss 287-8777
Consoer-Townsend and Assoc
1671 the Alameda
San Jose, Ca, 95126
Towill Inc 5 years ago
about 3000 ft and 1500-1800 ft with 6" lens
shot at 200 scale, mapped at 40 scale (1"=40 ft)
Jerry Sample 982-1758
$25/pair; $50 two pairs
delivered to Ray Mason, surveying dept of engineering at airport 876-2364
needs permission from airport for enlargements
Mr. R.G.Lee, 58-876-2224
P.O.Box 8097
San Francisco Int Airport
San Francisco, Ca, 94128
calculation of resolution of that series
6" focal length, 1500 ft altitude
altitude/focal length = field of view / image size
field of view at 1500 ft is 9"*1500/6"=2.7*10↑4"
resolution is approx 100 lines/mm * 225 mm = 2.25*10↑4 lines
resolution per line is 1.2" per pixel on ground
these others don't fly low-altitude
Santa Clara Aerial Surveys
2317 De La Cruz
Santa Clara, 248-3000
∂29-MAR-76 1551 FTP:Garvey at SRI-AI ftp
I have discovered that you can get files from here (and most TENEX sites)
by logging in as user ANONYMOUS, and using your initials as a password
and account. After that, you can get and retrieve any files that you would
normally have access to as a regular (unaffiliated) user. Try this for the
picture files and see how it works. If you still have problems, I'll see
what I can do from here.
Tom
∂29-MAR-76 1556 FTP:Garvey at SRI-AI FTP update
I found out a couple more facts about ANONYMOUS: first, it is a GET only
user name. I.e., you cannot STORE with it. Therefore, do not supply the
account field when logging in. Good luck,
Tom
∂03-Feb-77 2322 BLB via ECL images
Tom-
There are 20 image files on [1,blb] that I have ftp'd to you.
These are the long promised multi-wavelength images.
Could you do two things asap. First take a random sample
and see if they display correctly (they are in hand-eye
format). Second can you talk to Les or someone to make
sure they are not purged soon. Please let me know if all goes well.
I would like to know soon so I can clean them off my local (usc)
disk. If they are ok I will send you prints and info about what you
have in front of you.
Regards-
Bruce
∂11-Feb-77 2118 BLB via ECL images
I will have someone check to see if you got the headerless files
and send you new images.
bruce
PIX
∂04-Nov-76 2230 BLB via ECL pic files
I have transfered the two image files to [1,blb] that I mentioned
last week. They are called a1.hef and a2.hef. Please delete from
my area when you get your copy.
The guy that keeps track of the multisensory data is out of town
on the review so I have not been able to get thaat part to the
disk to send to you.
Will do asap.
-Bruce.
Tom-
I have FTP'd four of the cdc images to my area [1,blb].
They are cdca,......cdcd all with extension .hef
They should all be in hand eye format so you should be able to read them
without any problem. They are all of the "Blackdog" powerplant series.
Also they are all sequential in range (same camera angle but at 4 ranges).
Let me know if they are any interest. I can snd more.
I will also send you some of the multisensor stuff early next week.
Good luck-
Bruce.
∂10-FEB-76 1143 FTP:NEVATIA at USC-ECL digitizing
2000*2000 is probably the largest practical number of pixels obtainable
from our Optronix. This would require about 4-5 hours of digitizing
(larger areas require more time, even for the same number of
points). Color would take 3 times longer. We have not worked with
color in some time, but presumably it can be handled without much
hassle. Turnaround time is 1-2 weeks.
If you want some pictures digitized, please send me complete and
detailed info about the area, the resolution etc.
Haven't had time to work on the paper yet. I am going away
for one week, will get to it then.
NEVATIA
Digitizing your pictures will take about 3 weeks.
You have asked for digitizing some parts at 20 microns/pixel.
At this resolution, film grain noise can be a problem
(film grain is 2-5 microns across). If you do not need all the
resolution, it may be better to digitize with lager aperture.
Your color picture does not have too many colors (is mostly blue).
Do you really want it digitized in color?
I like your black and white photographs. Can you give sources
for us to acquire some photographs of similar terrain and quality?
I have been trying to find stereo pairs for myself, the problem
being to get low altitude pictures so that different depths can
be seen (your pictures do not seem to have much disparity either).
We do not have a standard database yet, but we are trying to build
one. Initially, it will consist of scenes of small numbers of
objects (such as one or two) at close distance. It is hard to find
pictures of just the right complexity.
Finally, the curved objects paper is ready to go out today; I
presume you have no objections.
Ram.
∂23-APR-76 1049 FTP:NEVATIA at USC-ECL
Date: 23 APR 1976 1043-PST
From: NEVATIA at USC-ECL
To: binford at SAIL
The files to copy are <JOHNSTON>AERIAL.STAN-1
and <JOHNSTON>AERIAL.STAN-2.
We called them A1.PIC and A2.PIC.
The pictures are in the USC format, currently I do not
have a program to convert to H/E format for 512*512
but may have one soon. If you need format info, leave
me a note.
Ram.
-------
∂26-APR-76 1030 FTP:JOHNSTON at USC-ECL DIGITIZED PICS
Date: 26 APR 1976 1030-PDT
From: JOHNSTON at USC-ECL
Subject: DIGITIZED PICS
To: BINFORD at SAIL
cc: JOHNSTON
TOM,
I AM SCOTT JOHNSTON. I WILL BE DIGITIZING YOUR PICS ON OUR
OPTRONICS HERE AT USC-IPI. I THINK YOU ALREADY HAVE THE FIRST
TWO PICS THAT WE HAVE DONE, BUT LET ME KNOW IF YOU DON'T, AND I'LL
PUT THEM BACK IN MY DIRECTORY. TODAY I FINISHED TWO MORE.
THEY ARE 512*512 SQUARE, STEREO PAIRS OF THE PARKING LOT.
IN MY DIRECTORY AS <JOHNSTON>OPTSC.512;1
AND <JOHNSTON>OPTSC.512;2
We called these A3A.PIC and A3B.pIC.
SCOTT JOHNSTON @USC-ECL
YES, THOSE ARE THE RIGHT PICS. AM NOW WORKING ON THE
512*1024 PICS.
SCOTT JOHNSTON USC-ECL
THE LAST OF THE BLACK AND WHITE NEGATIVES ARE DONE. THEY ARE
512 BY 1024 AND ARE IN MY DIRECTORY UNDER
<JOHNSTON>PIC-512-1024.PAIR;1 We called this S1.PIC
<JOHNSTON>PIC-512-1024.PAIR;2
SCOTT JOHNSTON USC-ECL
YOUR PIC IS NOW ON <JOHNSTON>STEREO.PIC;1
GOOD LUCK
SCOTT JOHNSTON USC-ECL
∂06-MAY-76 1122 FTP:NEVATIA at USC-ECL PICTURES
Date: 6 MAY 1976 1121-PDT
From: NEVATIA at USC-ECL
Subject: PICTURES
To: BINFORD at SAIL
Tom, we are still having trouble digitizing your color pic.
It will be a while before the problem is located and fixed
(several weeks). Do you wish to wait or me to return
your transparency? If you have color prints, we can digitize
them easily on the drum scanner.
Reminder: I would like some info on any stereo programs that
are in good enough shape for me to use here.
Ram.
-------
∂13-MAY-76 1059 FTP:JOHNSTON at USC-ECL DIGITIZED COLOR PICS
Date: 13 MAY 1976 1058-PDT
From: JOHNSTON at USC-ECL
Subject: DIGITIZED COLOR PICS
To: BINFORD at SAIL
TOM,
I'M VERY SORRY TO REPORT THAT WE ARE LIMITED IN SIZE FOR
COLOR SCANS. THE REASON THAT WE DIDN'T TELL YOU EARLIER
WAS THAT WE HAVE NEVER DONE COLOR SCANS THAT SMALL BEFORE,
AND WE ARE IN A DEVELOPMENTAL STAGE HERE. WE WERE TRYING TO
GET IT TO WORK OUT.
HOWEVER, WE CAN DO A COLOR SCAN FOR YOU IF YOU ARE WILLING TO
TAKE A LARGER AREA THAN YOU ASKED FOR. LET US KNOW.
ALSO, SORRY ABOUT THE INTERLACING OF THOSE B+W PICS. IT SEEMS
THAT WE USED A WRONG PARAMETER HERE. WE'RE WORKING ON THEM NOW.
SINCERELY,
SCOTT JOHNSTON @ USC-ECL
-------
∂14-MAY-76 1352 FTP:NEVATIA at USC-ECL DIGITIZING
Date: 14 MAY 1976 1342-PDT
From: NEVATIA at USC-ECL
Subject: DIGITIZING
To: BINFORD at SAIL
Gathered the following info on the digitizer:
1. Scanning is done continuously, but the light is turned
only for a short period where the sample is to be taken.
No integrating is performed. The detector is a photo-
multiplier.
2. The problem with the alternate lines is
due to amplifier(or sensor) rise time. Will try
to redigitize at slower speed.
3. Signal to Noise ratio is linear with the light received.
Ram.
-------
∂17-JUN-76 1103 FTP:JOHNSTON at USC-ECL STEREO PICS
Date: 17 JUN 1976 1057-PDT
From: JOHNSTON at USC-ECL
Subject: STEREO PICS
To: BINFORD at SAIL
TOM,
I HAVE TWO PAIRS OF PICS FOR YOU TO FTP.
THEY ARE:
<JOHNSTON>STEREO.PAIR;3,2 512x512
<JOHNSTON>STEREO1.PAIR;3,2 1024x512
THEY ARE 8-BIT, UNPACKED FORMAT. IF YOU DON'T GET THEM
TODAY, LET ME KNOW, AND I'LL PUT THEM UP AGAIN.
SCOTT JOHNSTON @ USC-ECL
-------
∂23-JUN-76 1408 FTP:JOHNSTON at USC-ECL PICS
Date: 23 JUN 1976 1409-PDT
From: JOHNSTON at USC-ECL
Subject: PICS
To: BINFORD at SAIL
TOMORROW JUNE 24 (THURSDAY) THERE WILL BE SIX PICTURES
IN MY DIRECTORY AFTER ABOUT 9:00 AM.
THEY WILL BE;
<JOHNSTON>STEREO.PAIR;1
<JOHNSTON>STEREO.PAIR;2 PACKED 512**2 8-BIT
<JOHNSTON>STEREO.BIG;1
<JOHNSTON>STEREO.BIG;2 PACKED 512**2 8-BIT
<JOHNSTON>STEREO1.PAIR;1
<JOHNSTON>STEREO1.PAIR;2 UNPACKED, 1024*512 LINES---8-BIT
LET ME KNOW IF YOU DON'T GET THEM.
SCOTT JOHNSTON @ USC-ECL
-------
∂24-JUN-76 1304 FTP:JOHNSTON at USC-ECL AERIAL PICS
Date: 24 JUN 1976 1305-PDT
From: JOHNSTON at USC-ECL
Subject: AERIAL PICS
To: BINFORD at SAIL
cc: JOHNSTON
ARE YOU GOING TO GET THOSE PICS TODAY???
6/24/76 THURSDAY
IF NOT, LET ME KNOW, AND I'LL PUT THEM UP AT A LATER DATE
WHEN YOU CAN GET THEM.
SCOTT JOHNSTON @ ECL
-------
∧
∂14-JUN-76 1458 FTP:JOHNSTON at USC-ECL PICS
Date: 14 JUN 1976 1457-PDT
From: JOHNSTON at USC-ECL
Subject: PICS
To: BINFORD at SU-AI
TOM,
YES THERE IS NEWS ON YOUR PICS.
WE ARE WORKING ON THEM, AND THEY WILL BE DONE SOON.
WE WERE HAVING A GREAT DEAL OF TROUBLE WITH THE INTERLACE PROBLEM,
BUT I THINK WE HAVE IT FIXED.
SCOTT
-------
∂17-JUN-76 1103 FTP:JOHNSTON at USC-ECL STEREO PICS
Date: 17 JUN 1976 1057-PDT
From: JOHNSTON at USC-ECL
Subject: STEREO PICS
To: BINFORD at SAIL
TOM,
I HAVE TWO PAIRS OF PICS FOR YOU TO FTP.
THEY ARE:
<JOHNSTON>STEREO.PAIR;3,2 512x512
<JOHNSTON>STEREO1.PAIR;3,2 1024x512
THEY ARE 8-BIT, UNPACKED FORMAT. IF YOU DON'T GET THEM
TODAY, LET ME KNOW, AND I'LL PUT THEM UP AGAIN.
SCOTT JOHNSTON @ USC-ECL
-------
∂23-JUN-76 1408 FTP:JOHNSTON at USC-ECL PICS
Date: 23 JUN 1976 1409-PDT
From: JOHNSTON at USC-ECL
Subject: PICS
To: BINFORD at SAIL
TOMORROW JUNE 24 (THURSDAY) THERE WILL BE SIX PICTURES
IN MY DIRECTORY AFTER ABOUT 9:00 AM.
THEY WILL BE;
<JOHNSTON>STEREO.PAIR;1
<JOHNSTON>STEREO.PAIR;2 PACKED 512**2 8-BIT
<JOHNSTON>STEREO.BIG;1
<JOHNSTON>STEREO.BIG;2 PACKED 512**2 8-BIT
<JOHNSTON>STEREO1.PAIR;1
<JOHNSTON>STEREO1.PAIR;2 UNPACKED, 1024*512 LINES---8-BIT
LET ME KNOW IF YOU DON'T GET THEM.
SCOTT JOHNSTON @ USC-ECL
-------
∂24-JUN-76 1304 FTP:JOHNSTON at USC-ECL AERIAL PICS
Date: 24 JUN 1976 1305-PDT
From: JOHNSTON at USC-ECL
Subject: AERIAL PICS
To: BINFORD at SAIL
cc: JOHNSTON
ARE YOU GOING TO GET THOSE PICS TODAY???
6/24/76 THURSDAY
IF NOT, LET ME KNOW, AND I'LL PUT THEM UP AT A LATER DATE
WHEN YOU CAN GET THEM.
SCOTT JOHNSTON @ ECL
-------
∧
∂28-JUN-76 1552 FTP:JOHNSTON at USC-ECL pics
Date: 28 JUN 1976 1553-PDT
From: JOHNSTON at USC-ECL
Subject: pics
To: binford at SAIL
tom,
I'll put up the picturees for you on wednesday. They"ll be in my
directory untill late that night. You'll be able to figure out the
titles because they'll be the only6 ones labelled stereo.
scott johnston @ usc-ecl
-------
∂15-JUL-76 1006 FTP:JOHNSTON at USC-ECL COLOR PICS
Date: 15 JUL 1976 1004-PDT
From: JOHNSTON at USC-ECL
Subject: COLOR PICS
To: BINFORD at SAIL
TOM,
We finally got the registration problems fixed on our
scanner. Right now we have the 512 x 512 scan ready for you
to ftp up. The files are:
<JOHNSTON>COLOR.RED;1
<JOHNSTON>COLOR.GRN;1
<JOHNSTON>COLOR.BLU;1
This data is eight bit, packed. We hope to get your
512 x 1024 in a couple of days.
Scott Johnston @ USC-ECL
-------
∂16-JUL-76 0841 FTP:JOHNSTON at USC-ECL AERIAL COLOR PICS
Date: 16 JUL 1976 0841-PDT
From: JOHNSTON at USC-ECL
Subject: AERIAL COLOR PICS
To: BINFORD at SAIL
TOM,
We now have all your pics ready for you. Let me know
if you didn't get the 512 x 512 pics yesterday, and I'll put
them up for you again. The pics we have for you today are
1024 pixels by 512 lines. Data is UNPACKED for these pics.
Let me know when you have them.
Scott Johnston @ USC-ECL
-------
∂16-JUL-76 0844 FTP:JOHNSTON at USC-ECL P.S.
Date: 16 JUL 1976 0844-PDT
From: JOHNSTON at USC-ECL
Subject: P.S.
To: BINFORD at SAIL
The files are:
<JOHNSTON>BIGCOLOR.RED
<JOHNSTON>BIGCOLOR.GRN
<JOHNSTON>BIGCOLOR.BLU
SRI
-------
∧
∂19-JUL-76 1322 FTP:JOHNSTON at USC-ECL PICS
Date: 19 JUL 1976 1323-PDT
From: JOHNSTON at USC-ECL
Subject: PICS
To: BINFORD at SAIL
TOM,
Your pics are in my dir now as:
<johnston>stan.red
<johnston>stan.grn
<johnston>stan.blu
They are 1024 pixels per line, 512 lines.
8 bit unpacked data.
Scott Johnston @USC-ECL
-------
20-JUL-76 1418 FTP:JOHNSTON at USC-ECL PICS
Date: 20 JUL 1976 1421-PDT
From: JOHNSTON at USC-ECL
Subject: PICS
To: BINFORD at SAIL
Tom,
Those pics are 1024 pixels by 512 lines (they scan faster
if we do them that way). When we scan anything over 512 pixels
wide, we have to write the data directly to mag-tape. Unfortunately,
the program only writes the data in unpacked form, so if you
ignore the alternate zeroes, your pic should be ok.
Scott Johnston @ USC-ECL
-------
∂28-JUL-76 1054 FTP:JOHNSTON at USC-ECL Tank, truck ...etc
Date: 28 JUL 1976 1055-PDT
From: JOHNSTON at USC-ECL
Subject: Tank, truck ...etc
To: BINFORD at SAIL
Tom,
I can put up the pics today if you want...let me know
if you can do the ftp today.
Scott Johnston @ USC-ECL
-------
∂28-JUL-76 1201 FTP:JOHNSTON at USC-ECL PICS
Date: 28 JUL 1976 1201-PDT
From: JOHNSTON at USC-ECL
Subject: PICS
To: BINFORD at SAIL
Tom,
The pics are up in my dir now...there is a coding scheme
that i think will help you to decide what to ftp.
There are a set called TANK.*;*
THE EXTENSION IS WHETHER IT IS A POSITIVE OR A NEGATIVE.
EXTENSIONS AND VERSIONS DO CORRESPOND TO EACH OTHER.
THERE IS ALSO A SET CALLED KLEIN.*;*
THEY ARE ALSO A SET OF POSITIVE AND NEGATIVES.
HOWEVER, THE POSITIVES ARE IDENTIFIED BY THE FIRST LETTER IN THE
EXTENSION. IF IT IS A 'P' IT IS POSITIVE...OTHERWISE NEGATIVE.
THE REST OF THE EXTENSION CORRESPONDS BETWEEN P/N PICS.
LET ME KNOW IF YOU DON'T GET IT ALL.
SCOTT JOHNSTON
-------
∂02-AUG-76 2044 BLB via ECL
Tom-
I will try to find out the best was to get you some pictures
so you can choose what you want. Will get back soon.
Sorry to take so long in answering your mail I have not been on
this machine for awhile. I am trying to get alot of backlog writting
out of the way so I can study for my oral comprehensive next quarter.
I want to get it out of the way so I can concentrate on something for a thesis.
KThere are several new reports that are either out or about to be out.
You should have already received one on control ideas. The next one
you should get is longer and has recent results and background on the
ARPA/STO contract. Although we have some much better results
now, I think you have seen most of what is in this paper. It will be mailed
as soon as the publication clearance comes through from ARPA's monitop
RADC.
The ccd people have come up with some really cleavor stuff. It will probably
be in usc's next report. If they write anything up before that I will
see that you get a copy.
How are things with you? Still enjoying married life I hope.
I am sort of out of things right now trying to get this test out of
way for school. I will try to get up to see you when I get that over.
Will get back to you as soon as I can about images.
∂02-AUG-76 2108 FTP:BULLOCK at USC-ECL cdc images
Date: 2 AUG 1976 2107-PDT
From: BULLOCK at USC-ECL
Subject: cdc images
To: jq, dudani
cc: tob at SAIL
How many images do we currently have transfered from the CDC tapes
to HEF format? How many of those do we have prints of from cdc (is that
the big sheet negative?).
How many 3 band images do we have from our own system (the bldg 21, or
what ever it is called, for example)? Are there good prints (optical, not
off the conographics) available?
Also, JQ , did you get the info from Mancill to read that
SAR image?
Please let me know as soon as you can, and if the numbers are large
just approximately, do not count. Tom Binford at Stanford has talked
to us about getting some of these for their ARPA contract work. He has
also talked to cdc, but we have stuff already in a form he can use.
If you let me know I will take care of getting some photo prints made
and sent to him. He wants to choose only a few, so it should
not be the big job making making many tapes, etc. as in the past
for other contractors.
Thanks- Bruce.
-------
∂28-MAY-76 0558 FTP:FELDMAN at SUMEX-AIM ARPA NETWORK IMAGE PROTOCOL
Date: 28 MAY 1976 0556-PDT
From: FELDMAN at SUMEX-AIM
Subject: ARPA NETWORK IMAGE PROTOCOL
To: BINFORD at SU-AI, NEVATIA at USC-ISI, TENNENBAUM at SRI-AI
cc: CARLSTROM at USC-ISI
THERE IS SOME WORK ON IMPLEMENTING THE NETWORK GRAPHICS PROTOCOL
BY DANNY COHEN AT ISI. THERE IS A QUESTION WHETHER IMAGE TRANSFER
SHOULD BE INCLUDED AND,IF SO, HOW. DANNY, BOB SPROULL AND I ARE GETTING
TOGETHER IN EARLY JUNE TO TAKE A FIRST LOOK AT THE PROBLEM. ANY IDEAS,
OPINIONS,ETC. THAT YOU CAN GET TO ME SOON ARE LIKELY TO HE;P US A LOT
IN AVOIDING THE GROSSEST ERRORS.
-------
∂09-JUN-76 2031 FTP:NEVATIA at USC-ECL NETWORK GRAPHICS
Date: 9 JUN 1976 1733-PDT
From: NEVATIA at USC-ECL
Subject: NETWORK GRAPHICS
To: feldman at SUMEX-AIM
cc: BINFORD at SAIL, TENENBAUM at SRI-AI
(Jerry, I am at USC-ECL , not ISI. Just recd your mail on june 9).
I have no strong feelings for including images in a graphics protocol
except for completeness. However, if a protocol is planned, obviously
we have to agree on a format. A header describing the format, such
as packing of pixels and the size of picture is essential.
Stanford format may be a good starting point, with an additional
record for describing the history of the picture (what has been done
to it).
Some devices (or sites) may not be able to work with variable
size pictures or packing formats, and some convention is needed
for calling pre-formatted pictures.
Finally, do half-tone image transfers qualify as images or as
graphics?
Ram Nevatia.
-------
∂10-JUN-76 0550 FTP:FELDMAN at SUMEX-AIM NETWORK IMAGE PROTOCOL
Date: 10 JUN 1976 0550-PDT
From: FELDMAN at SUMEX-AIM
Subject: NETWORK IMAGE PROTOCOL
To: NEVATIA at USC-ECL
cc: BINFORD at SAIL, TENENBAUM at SRI-AI
THE CURRENT THOUGHT IS THAT THE PROTOCOL WILL BE AS MUCH CONCERNED
WITH TALKING ABOUT IMAGES AS WITH SENDING THEM. THE IDEA IS TO DEFINE
A "RASTER" FACILITY TO BE ADDED TO THE GRAPHICS PROTOCOL. THIS
IS BEING COORDINATED BY ONE OF OUR STUDENTS, JOE MALESON, WHO IS
SPENDING THE SUMMER WITH SPROULL @ PARC. PLEASE CONVEY THOUHTS ON THE
MATTER TO MALESON@PARC AS WELL ( WHICH I JUST FORGOT TO DO).
-------
∂11-JUN-76 2156 FTP:Tenenbaum at SRI-AI Network Image Protocol
Date: 11 JUN 1976 1513-PDT
From: Tenenbaum at SRI-AI
Subject: Network Image Protocol
To: Feldman at SUMEX-AIM
cc: Nevatia at USC-ECL, Binford at SAIL, Maleson at PARC
Just got the June 1 Message. Regarding a format for transmitting
images, the format is relatively unimportant if you know what it is.
Therefore, we concur with Nevatia in emphasizing the need for a
standardized header which describes the format of what follows.
Regarding Talking about images, we have found a simple pixel
level description useful: associated with each pixel is a number.
A correct partitioning of the scene into regions could then be represented
by assigning the same (unique) number to all pixels that belong in the
same region. The desired partition could then be recovered with a trivial
connectivity analysis and encoded in the users favorite data structure.
The number coding also provides an easy way of transmitting the
output of someone elses low level segmentation program for use by
a high level interpretation program. (The interpretation program constructs
its own favorite datastructures from the connectivity analysis.)
A legend can also be included, associating with each number a semantic
interpretation (or set of possible interpretations). This allows an
interpretation program to test its own results, do learning etc.
-------
∂15-AUG-76 2142 BLB via ECL images
Tom-
Progress report on your image request. I was going to make some xerox
copies of some poor reproductions of imagery you could get. I then found
that there is a final report now in print that has good quality prints.
Decided it would be better for you to wait. So will send asap.
I noticed the mail note about MACLISP. I have heard that it has very good
arithmetic (as good or better than FORTRAN). Would you vote for
or against use in scene analysis, as an attempt to get both good
arithmetic efficiency and good symbolic language all in one
package?
Talk to you soon-
Bruce.
∂15-AUG-76 2149 BLB via ECL
My mail note may not have been clear. By "in print" I meant that it
is in the repro dept. queue. So will not be available until it is
printed.
Sorry for lack of detail.
∂11-MAR-76 2158 BLB via ECL
Sorry to be so slow in answering you message.
I am interested in getting together on both the hardware
issue and the terminal homing problem. I would like to get
together with you sometime fairly soon. Are you pln↓anning on
being out of town in the next few weeks? I have not talked to ram yet, but
I suppose it would be most efficient if he and I got together first in
L.A.
By the way, a paper will be completed sometime in the next month on the
current status of our terminal homing approach (although it is not called
that in the paper). We have really learned a great deal about the problem.
The first cut system will be of no surprise to you technically, except perhaps
for the fact that it works.