Image Size Reduction
Ex. lan | 786432 0
Ex9% 84365 702067
Ex11% 71090 715342
Ex12% 65592 720840
Ex13% 59345 727087
Ex14% 55284 731148
Table 4. Amount of Reduction in Storage for
Experiement Ex (in bytes)
5. CONCLUSION
The problem of compressing images that are
used for mapping using JPEG is of a
compromising nature, since more compression
saves a significant amount of space and
facilitates image processing but causes
appreciable loss of information. For some
applications of digital mapping, compression
might be very useful even if some information
is lost. In this experiment, compressing the
image up to 12%, is possible with little effects
on the visual appearance and with appreciable
changes in the pixels number of some classes.
In this particular experiement, for a three-band
TM image, more than 700 kb of storage could
be saved in the compression process. If high
accuracy is not necessary for identifying image
features, the classification parameters can be
adjusted to accommodate the minor changes in
the pixels values. It is also noticed that the
effect of JEPG is a scene-dependent matter. It
seems that scenes of few heterogeneous classes
may be more amenable to high compression rate
than scenes with many homogeneous classes.
ACKNOWLEDGEMENT
The author would like to thank King Abdul Aziz
City for Science and Technology (KACST) for
provideing the TM data for this study.
28
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B3. Vienna 1996
REFERENCES
Congalton, G. R., 1991. A review of assessing
the accuracy of classifications of remotely
sensed data. Remote Sensing of Environment,
37(1), pp. 35-46.
Miller, S., U. V. Helava, and K. D. Helava,
1992. Softcopy photogrammetry workstations.
Photogrammetric Engineering & Remote
Sensing, 58(1), pp. 77-83.
Jensen J., 1986. Introductory Digial Image
Processing: A Remote Sensing Perspective.
Prentice-Hall, New Jersey, U.S.A., pp.
177-230.
Lammi, j., and T. Sarjakoski, 1995. Image
compression by the JPEG algorithm.
Photogrammetric engineering & remote sensing,
61(10), pp. 1261-1266.
Langdon, Gulati, G. A. and E. Seiler, 1992. On
the JPEG model for lossless image compression.
In: Proceedings of Data Compression
Conference (DCC’92), IEEE Computer Society,
pp. 172-180.
Paik, H. W., and A. Khubchandani, 1992.
Quantization scheme for JPEG baseline
sequential encoding of still images. IEEE, pp.
976-979.
Pennebaker, W. B., and J. Mitchell, 1988.
Probability estimation for the Q-coder. IBM J.
Res. Development, 22(6).
Wallace, G. K., 1992. The JPEG still picture
compression standard. IEEE Transaction on
Consumer Electronics, 38(1), pp. 18-34.
ABST
Ane
coding
edge li
Line:
the mai
Edge-
represei
togethe
age has
edges o
a select
ing to s
In ori
comput
is usual
descrip
tions w
tation
(Green
+-s- rep
tion an
cused |
Line
since t
proxin
(Gonz:
Line
edges |
the alg
extract
detecti
es to li
2.1. F
The pr
applie
basic 1
image
Figure
The ce