Title: Elements of Photographic Systems Chapter 2, Remote Sensing and Image Interpretation
1Elements of Photographic SystemsChapter 2,
Remote Sensing and Image Interpretation
2Topics
- Digital Imaging
- Image Capture
- Color Digitization
- Color Corrections
- Image Storage
- Aerial Videography
- Multiband Imaging
3Electronic Imaging(Khosla, 1992)
- Data Capture Photosensitive solid state devices
- Data storage Magnetic, optical, and solid-state
media - Data manipulation Digital image processing
- Data transmission Telemetry, telephone lines,
computer networks - Softcopy display Computer monitors, television,
projection video - Hardcopy display Dye sublimation, inkjet,
thermal, and laser printers
4Digital (Electronic) Image Fundamentals
- The digital image
- Capture
- Processing
- Storage
- Output
5Image Enhancement
(Kodak)
Multiply or divide each pixel by constant value
Reverse image so black becomes white
Add or subtract constant value
6Analog vs. Digital Curves
- Analog (film) image
- Continuous analog waveform
- Digital image
- Digital values derived from sampling analog image
- Discrete electronic pulses translated into
strings of zeros and ones
(Kodak)
7Digital Fidelity
(Kodak)
Vertical and horizontal cells sampled at same time
8 bit converter represents 28 256 brightness
levels, 12 bit 212 represents over 4000
Pixel has red level of 227, green level of 166,
and blue level of 97
8The Digital Image (Vector and Raster Graphics)
- Vector (object-oriented) graphics
- Stored as display list giving location and
properties of objects - Shapes, arcs, and lines
- Takes a few kilobytes storage
- Raster (bit-mapped) graphics
- Stores location / value of each pixel
- Can take thousands of times more memory than
vector
9The Digital Image (Vector and Raster Graphics)
(Kodak)
10Image Capture
(Kodak)
Digital image system employs CCD as detector
Basic system is lens film
Pixels created by sensing light intensity of
small portion of film
11Color Digitization
(Kodak)
12Image Quality
(Kodak)
Quality of raster image determined by
spatial resolution and brightness resolution
Pixel resolution determined by rate at which
scanner samples
Brightness or color value of each pixel defined
by one bit or group of bits
13The 24-Bit Pixels
(Kodak)
(16.7 million colors)
(3/ 8-bit sets of numbers)
(one of 256 brightness values for red, green, and
blue)
14Color Corrections
(Kodak)
Reduce green by 50
Reduce red by 50
Reduce blue by 50
15Image Storage
(Kodak)
One 35 mm negative would require 18 megabyte file
Same negative would require 14 high
density floppy disks
120 megabyte hard drive could store six of these
images
16Aerial Videography
- Form of electronic imaging whereby analog or
digital video signals are recorded on magnetic
tape or magnetic or optical disks - Cameras for aerial videography can have many
configurations - Single-band
- Multiband
- Multiple single-band
- Cameras can sense in some or all of the
following - Visible
- Near-IR
- Mid-IR
- Following configurations are available
- Shuttered cameras--exposure times as short as
1/10000--used for both analog and digital aerial
videography - Analog video recording with 485 horizontal lines
per image frame - Digital video recording with horizontal
resolution of 500 TV lines (TVLs)
17Aerial Videography (Concluded)
- Review of plate 6 (after page 20)
- (a) and (b) are images of cotton field with
harvester ants - Spatial resolution of video image (a) lower than
photograph (b) - Most of the ant mounds can easily be detected in
(a) - (c) and (d) are images of a cotton field with
saline soils - Barren to sparsely vegetated saline areas as
readily distinguished in (c) as in (d) - Color tones of video composites comparable to
color IR film tones in both scenes
18Multiband Imaging
- Both normal color and color IR films are
three-layer systems - Normal color film layers individually sensitive
to blue, green, and red energy - Color IR film--used with appropriate filter-- has
layers individually sensitive to green, red, and
near-IR energy - Three-layer films produce color images because of
dye layers that control amount of blue, green,
and red energy that form the final photographic
image - Best combination of multiband images for
discriminating a given scene varies with spectral
response pattern for objects of interest in the
scene - Understanding of object reflectances (and/or
emittances) and application to multiband imaging
normally enhances contrast between different
terrain feature types and between different
conditions of the same feature type - Plate 7 (after page 20) shows normal color and
color IR images that were prepared from
individual band images shown in table 2.3, pg 122
19Four Views of Crab Nebula from Different
Multispectral Sensing Devices
X-ray
Optical
(rst)
Infrared
Radio
20Supplemental References
- NASA Goddard Remote Sensing Tutorial,
http//rst.gsfc.nasa.gov/intro/Part2-12.htm - Lenses The Basics, http//cybercollege.com/typo11
.htm - Kodak Digital Learning Center, http//www.kodak.co
m - Image Information in the Eye and the Camera,
http//www.dai.ed.ac.uk/CVonline/LOCAL_COPIES/OWEN
S/LECT1/node2.htmlSECTION00022100000000000000 - Photogrammetry, http//www.sli.unimelb.edu.au/nico
le/lectures/lecintro.html