Presented by M. Malini Deepika, Research Scholar, SASTRA

Presented by
M. Malini Deepika,
Research Scholar,
SASTRA University.
Thanjavur. Tamilnadu.

STEREO Earth Observers
- Solar TErrestrial Relations Observatory
- Two identical space cameras are launched
- Single space craft carries two earth observing sensor

Need for Stereo Earth Observers
- Information of depth and width of the ravination

Difficulty Behind Stereo Earth Observers
- upholding the craft
- Limitation in sensor
- Life time of the Craft

DLR- 3D Line Stereo scanner (1980)
- offered to fly on SROSS-I (1988)
- derive coarse Digital Surface Models (DSM)
- 70 m ground resolution
- German 3-line scanner MOMS-02

HRS Instrument on SPOT-5
- MOMS-02/D2 (1993)
-MOMS-2P(1996-1999)
- MOMS-03
- Special additions to S/w systems (stereo IKONOS-2 and
QuickBird images )

Cartosat-I (2005)
-E.Baltsaviaset.al., et.al. (2007)
- Willneff. et.al., and Y.V.N. Murthy et.al (2008)
- Giribabu (2014)

Dual Camera Based

Depth Camera Based

Single Camera Based
- Consecutive frame based
methods
- Depth estimation based method

Hardware Complexity
- Multiple camera
- Multidimensional View

Color distortions

Retinal rivalry

Ghosting

Computational Intensity
- Depth Estimation Based Methods
Foreshortening
Specular Surfaces
Image Courtesy ISPRS

To get anaglyph video
sequence in real time

Using Monocular Vision




Real world projection of the object
Estimating depth of the object
Sensor Independent
Data from any period can be used
Step: 1 Acquire the scene
Step: 2 Stereo pair is generated
(m) The coordinates of Clone1 frame is defines as,
A( x A, y A, z A )  ( x A  b , y A , z A )
2
(n) The coordinates of Clone2 frame is defines as,

Ax A' , y A' , z A'   x A'  b , y A' , z A'
2

Step: 3 Colour component of the cloned frames A and A’ are changed as red
and cyan
Step: 4 The colour component converted cloned frame are superimposed to
form the projection on the single frame. The projection is done by
Euclidean geometry
xA

x

x A'

x

A
b
2
f
ZA
A'
b
z A'
 Step: 5 The anaglyph video is generated
2
f

Agriculture

Medical
Defence



Arts and Entertainment
Virtual office
Chi-Square test on Contrast in unprocessed PAN Image
Distortion
GN
Motion
SN
Chi Value
0.295
0.4193
0.3987
p
0.001
0.009
0.007
Chi-Square test on Contrast in Anaglyph PAN Image
Distortion
GN
Chi Value
p
0.2677
0.008
GN-Gaussian Noise
SN- Salt Noise
Motion
SN
0.422
0.3498
0.004
0.003
Chi-Square test on Contrast in source Multispectral Image
Distortion
GN
Motion
SN
Chi value
0.3294
0.4219
0.3948
P
0.002
0.008
0.002
Chi-Square test on Contrast in Anaglyph MS Image
Distortion
GN
Motion
SN
Chi value
0.2850
0.4439
0.3690
P
0.000
0.011
0.009
Advantages
Easy
Less
Any
implementation
hardware complexity
type of anaglyph image
can be generated
No
image stabilization or
rectification is required

Giribabu D., Pramod Kumar, John Mathew, K.P. Sharma and Y.V.N. Krishna Murthy,2013.
DEM generation using Cartosat-1 stereo data issues and complexities in Himalayan terrain in
European Journal of Remote Sensing ,46,pp. 431-443.

Hae-Yeoun Lee, Taejung Kim, Wonkyu Park, Heung Kyu Lee ,2003. Extraction of digital
elevation models from satellite stereo images through stereo matching based on epipolarity
and scene geometry in Image and Vision Computing, 21,pp. 789-796.

Sreenivas Kandrika, R. S. Dwivedi,2013. Reclamative grouping of ravines using Cartosat-1
PAN Stereo Data in J Indian Soc Remote Sens , Sci Vis 41 (3),pp. 731-737.