Project Title:

A Real-Time Image compression System for Use in Autonomous

Q-DOT Photonics, Inc.
2864 South Circle Drive, Suite 900
Colorado Springs, CO 80906

93-1-07     09   7803

A Real-Time Image compression System for Use in Autonomous
Navigation

Abstract:

High compression ratios possible through wavelet transforms offer
the potential for realtime image transmission over narrow-band
data channels. This type of transmission capability would allow
for low-power image transmission using existing communication
hardware. Q-DOT proposes a real-time image compression system
based upon optically based wavelet decomposition. Optical
processing allows for massively parallel processing of the image
data. A novel electro-optic spatial light modulator is used to
preprocess the image by programmably correlating the input image
with a wavelet basis set. The image compression system is
programmable allowing the wavelet basis (resolution set) to be 
optimally selected based upon the operating scenario. The system
proposed is independent of the type of image data and is
therefore compatible with a variety of image types including
LADAR, conventional visible light, infrared and even RADAR data.
The Phase I objective is to analyze the technical feasibility of
the proposed image compression system including prototyping an 
electro-optical modulator cell which is key to the compression
concept. The experimental and theoretical data gathered during
Phase I will provide the foundation needed for a Phase II
development effort.

The proposed system has applications in medical imaging where it
is desirable to isolate features of particular size or texture
(wavelet decomposition). This type of feature enhancement is used
in mammographic cancer detection, bone fracture identification,
and osteoporosis studies. Other application exist in the area of
multimedia communications including teleconferencing and High
Definition Television (HDTV) where the ability to rapidly 
compress and decompress images for narrow band transmission is of
great interest.

Wavelet, image-compression, imaging, transform, real-time,
electro-optic.