3 Processing Remotely Sensed Data: Hardware and Software Considerations
47
language. The programming languages most often used in the development of
digital image processing software are Assembler, C, JAVA, and FORTRAN.
Many digital image processing systems provide a toolkit that programmers can
use to compile their own digital image processing algorithms. The toolkit consists
of primitive subroutines. such as reading a line of image data into RAM or writing
a vector to the screen. Also many software packages provide high-level macro
languages, or libraries of image processing and GIS software components. These
tools allow the development of complex applications for specific purposes.
3.1.5 Mass Storage
Digital remote sensing research often requires substantial mass storage resources.
For example, during the early part of a project it is not uncommon to load an entire 7-band Landsat TM scene onto a hard disk at one time (3,000 x 3,000 pixels x
7 bands = 63 MB) to obtain an appreciation of the geographic extent of the image
and its apparent quality. Therefore, the storage media should have rapid access
times, have longevity (i.e. last for a long time), and be inexpensive (Rothenberg,
1995). Digital remote sensor data (and other ancillary raster GIS data) are normally stored in a matrix band sequential (BSQ) format in which each spectral
band of imagery (or GIS data) is stored as an individual file. Each picture element
of each band is represented in the computer by a single 8-bit byte (with values
from 0 to 255). The best way to make the brightness value available to the computer rapidly is to place the data on a hard (or optical) disk where each pixel of the
data matrix may be accessed at random and at great speed (within microseconds).
The cost of hard disk storage per gigabyte continues to decline rapidly. It is common for digital image processing laboratories to have gigabytes of mass storage
associated with each workstation as suggested in Fig. 3.1. In fact, many image
processing laboratories now use RAID (redundant arrays of inexpensive disks)
technology in which two or more drives working together provide increased performance and various levels of error recovery and fault tolerance (Freedman,
1995).
Figure 3.4 depicts several types of analog and digital remote sensor data mass
storage devices and the average time to physical obsolescence, i.e. when the media
begin to deteriorate and information is lost. Properly exposed, washed, and fixed
black and white aerial negatives have considerable longevity, often more than 100
years. Color negatives with their respective dye layers have longevity, but not as
long as the black and white negatives. Similarly, black and white paper prints have
greater longevity than color prints (EDC, 1995; Kodak, 1995). Hard and floppy
magnetic disks have relatively short longevity, often less than 20 years. Magnetic
tape media (e.g. 3/4" tape, 8-mm tape, and 112" tape shown in Fig. 3.4) can become \lnreadable within 10 to 15 years if not rewound and properly stored in a
cool, dry environment (EDC, 1995).
Optical disks can now be written to, read, and written over again at very high
speeds (Normile, 1996). The technology used in re-writeable optical systems is
magneto-optics (MO), where data is recorded magnetically like disks and tapes,
but the bits are much smaller because a laser is used to etch the bit. The laser heats
47
language. The programming languages most often used in the development of
digital image processing software are Assembler, C, JAVA, and FORTRAN.
Many digital image processing systems provide a toolkit that programmers can
use to compile their own digital image processing algorithms. The toolkit consists
of primitive subroutines. such as reading a line of image data into RAM or writing
a vector to the screen. Also many software packages provide high-level macro
languages, or libraries of image processing and GIS software components. These
tools allow the development of complex applications for specific purposes.
3.1.5 Mass Storage
Digital remote sensing research often requires substantial mass storage resources.
For example, during the early part of a project it is not uncommon to load an entire 7-band Landsat TM scene onto a hard disk at one time (3,000 x 3,000 pixels x
7 bands = 63 MB) to obtain an appreciation of the geographic extent of the image
and its apparent quality. Therefore, the storage media should have rapid access
times, have longevity (i.e. last for a long time), and be inexpensive (Rothenberg,
1995). Digital remote sensor data (and other ancillary raster GIS data) are normally stored in a matrix band sequential (BSQ) format in which each spectral
band of imagery (or GIS data) is stored as an individual file. Each picture element
of each band is represented in the computer by a single 8-bit byte (with values
from 0 to 255). The best way to make the brightness value available to the computer rapidly is to place the data on a hard (or optical) disk where each pixel of the
data matrix may be accessed at random and at great speed (within microseconds).
The cost of hard disk storage per gigabyte continues to decline rapidly. It is common for digital image processing laboratories to have gigabytes of mass storage
associated with each workstation as suggested in Fig. 3.1. In fact, many image
processing laboratories now use RAID (redundant arrays of inexpensive disks)
technology in which two or more drives working together provide increased performance and various levels of error recovery and fault tolerance (Freedman,
1995).
Figure 3.4 depicts several types of analog and digital remote sensor data mass
storage devices and the average time to physical obsolescence, i.e. when the media
begin to deteriorate and information is lost. Properly exposed, washed, and fixed
black and white aerial negatives have considerable longevity, often more than 100
years. Color negatives with their respective dye layers have longevity, but not as
long as the black and white negatives. Similarly, black and white paper prints have
greater longevity than color prints (EDC, 1995; Kodak, 1995). Hard and floppy
magnetic disks have relatively short longevity, often less than 20 years. Magnetic
tape media (e.g. 3/4" tape, 8-mm tape, and 112" tape shown in Fig. 3.4) can become \lnreadable within 10 to 15 years if not rewound and properly stored in a
cool, dry environment (EDC, 1995).
Optical disks can now be written to, read, and written over again at very high
speeds (Normile, 1996). The technology used in re-writeable optical systems is
magneto-optics (MO), where data is recorded magnetically like disks and tapes,
but the bits are much smaller because a laser is used to etch the bit. The laser heats