EROS DATA CENTER DENSITY CONTROL SCALE
Intensity values 0 8 17 25 34 42 51 59 68 76 85 93 102 110 119 127
16-step scale _ 0 1 2 3 4 5 6 7 8 9 10.11 12 13 14 1 15
9,200 — Shadow
endpoint
1.00
.80
.70
Three-
75—60 quarter
tone
o
po u.50
2. = o
uJ o +
P £4 zZ
um ©
= =
| O —-
e atone §
50—.30 Midtone Oo
uw. W 2
$ 6 :
E L =
du 3
I 1.20 ¥
cg
<
& 255 Quarter
© ©
uw tone
+
> z 10
TN Highlight
5—02 endpoint
0
.08 .16 .26 34 43 51 60 69 .80 .89 .99 1.08 1.19 1.30 1.42 1.52
ORIGINAL CONTINUOUS-TONE FILM DENSITIES
Figure 3.--Halftone reproduction curves of the Washington, D.C., and
vicinity TM Landsat image map (USGS, 1984).
and magenta, and cyan. The approximate differences are shown in the
C.I.E. (Commission International d'Eclairage) diagram (fig. 4). The color
hue and saturation attained on an electronic color display or a photo-
graphic color proof can be made to closely approximate the final litho-
graphic print by linearizing the output of the color photographic proof
described in the digital mosaic procedure (fig. 1).
Photomechanical Mosaic Procedure
Film images from the LBR and paper prints of each Landsat scene to be
photomechanically mosaicked are prepared and are evalueted for uniformity
in image detail and range of contrast.
Ground control points are selected from 7.5-minute topographic quadrangles
at 1:24,000 scale; control and tie points are transferred to band 5 images,
then to corresponding images of bands 4 and 7 for a typical MSS Landsat
project.
Control points are measured and each band 5 image is adjusted to ground
control using a least-squares similarity transformation. All band 5
images are then adjusted with a simultaneous block adjustment.
A control base sheet is plotted at publication scale. The band 5 film
images are enlarged to fit the control base, matching densities in image
overlap areas; this process is repeated for bands 4 and 7.
211