Page 1
GE
Intelligent Platforms
applied image
processing
Page 2
The key to an image processing
solution is what you can’t see.
The peace of mind and security you gain from
working with a proven supplier is invaluable.
Image processing solutions from
GE Intelligent Platforms are deployed in
a wide range of military and paramilitary systems worldwide, including UAVs,
air-to-ground, ground-to-air, ground-toground, naval tracking, remote control gun
platforms and range instrumentation.
Our rolling program of research and
development ensures that your image
processing solutions remain at the forefront
of the technology. Whether it is advanced
algorithms to provide automated refueling
of unmanned aerial vehicles, ultra miniature electronics to enable implementation
in infantry weapon sights, or increased
processing power to handle high resolution
sensors, our research ensures the solution
is available to meet the challenge.
We are world leaders in the design and
manufacture of rugged, real-time image
processing subsystems, and one of the
leading independent suppliers of ‘commercial-off-the-shelf’ video tracking and image
processing systems to the global defense,
aerospace and security market.
The company is approved to ISO 9001:
2000 and AS/EN 9100 for design and
manufacture. The company is committed
to long-term support of its customers,
many programs running for up to 20 years,
and a full Product Lifecycle Management
package is available.
Applied imAge processing
Page 3
Our experience includes
the full range of applications.
Whether you have a ground, sea, or air application (or a
combination), we have the kind of experience you need.
Developing and deploying an advanced
electro optical system can be a difficult process. So it’s comforting to know that you can
turn to us with the assurance that we can
provide cutting-edge image processing, plus
the domain expertise to support the development of your system through deployment
and throughout its life. Whether you require
an individual board or a fully integrated system, we have untold hours of engineering
experience designing, testing and validating
rugged image processing solutions.
For over two decades, GE Intelligent
Platforms has developed a wide range of
real-time video image processing solutions to automatically extract the maximum amount of useful information from
raw video data and present it to users in
an easily digestible form so that system
integrators can achieve superior electro
optical system performance.
Each of these solutions is designed to
provide optimum performance in compact, low power, environmentally-proven
hardware packages. Most of our image
processing solutions use standard ‘off-theshelf’ products that meet interface and
performance requirements without the
need for major modification.
Image processing can be grouped into
two areas ‘Data Extraction’, gathering
intelligence from the video image and
‘Visualization’, processing the video image
for easier understanding by the operator.
Page 4
Data Extraction – Target
Detection and Tracking.
The detection and tracking of multiple small
targets in high clutter environments.
Modern warfare on land, sea and in the air
demands the capability to detect, classify
and engage threats at extended ranges.
Electro optical systems make a significant
contribution to meeting this demand,
and image processing provides a major
performance-enhancing element.
Whether it is to increase system accuracy
and reduce operator workload for manned
systems, or facilitate autonomous operation of unmanned vehicles, GE Intelligent
Platforms can provide an effective image
processing solution.
The GE Intelligent Platforms ADEPT family of image processing modules has been
optimized to perform real-time target
detection and tracking across the full range
of military electro optical system applications. Whether integrated in air-to-ground,
ground-to-ground, ground-to-air or naval
electro optical systems, the algorithms at
the heart of the ADEPT family provide effective detection and tracking against the full
spectrum of targets.
Automatic video tracking involves almost
all image processing functionalities: from
image enhancement and target detection/
acquisition, to target tracking and system
support functions.
Applied imAge processing
Page 5
Our video tracker hardware and proven algorithms are readily adaptable to your application.
Image Enhancement
The source video is fed to a pre-processor
which uses verified mathematical algorithms to accomplish image enhancement,
improve target contrast and suppress
clutter, providing a more accurate separation of the target from ‘noise’ in the scene
and more robust target acquisition and
tracking. A range of different pre-processors are included as standard.
Target Detection/Acquisition
The system then identifies objects within
an area or window of the video image
which meet the user-defined target
criteria. A range of different detection
algorithms are built in to meet different
situational requirements.
When one or more targets have been
detected, the system can enter track mode,
either automatically or under operator control. Automatic acquisition may be prioritized
using several different factors, including
target nearest boresight and largest target.
Algorithms
The GE Intelligent Platforms automatic
video trackers feature a wide range of
proven high-performance detection and
tracking algorithms which can be tailored
for the particular operational scenario,
including centroid, phase correlation and
edge. Algorithms may be combined to meet
particularly demanding tracking scenarios.
Features
Automatic video trackers from GE Intelligent
Platforms offer a range of communication
interfaces, video formats, platform PID filters,
joystick shaping, graphics overlay planes,
and field of view compensation. Tracker and
system set up may be optimized via Micro-
®
soft
Windows®-based software.
Target Tracking
In Autotrack Mode, the video tracker automatically tracks the selected target and can
control almost any type of pan and tilt or
gimbal system to keep the target centered
in the field of view.
Page 6
VISUALIZATION
Image Stabilization.
Enhancing operational performance and reducing
fatigue by providing a clear, stable image.
Real-Time Image Stabilization
Many surveillance and targeting systems,
particularly those with narrow fields of
view, suffer from vibrations that can lead to
unstable imagery being presented to the operator. These unwanted camera movements
can be a huge obstacle to interpreting the
video image, resulting in a significant detrimental effect on operator efficiency. While
it is possible to remove this motion using
mechanical inertial stabilization methods,
this comes at the expense of size, weight,
power consumption and cost, all of which
can affect the viability of a program.
The GE Intelligent Platforms real-time
image stabilization system will electronically remove unwanted global motion from
a video stream. It uses our SceneLock
algorithm to detect image rotations, scaling and translations by tracking a large
number of points in the image and then
combining the motion estimates of each
point to determine the overall scene motion from one video frame to the next.
Commands can be issued to the camera
platform to pan and tilt based on this motion
measurement in such a way as to keep the
aimpoint at the centre of the image. In many
cases, this results in a stable image. In some
*
situations, the camera platform cannot take
out all of the motion due to data transmission and processing latencies or the frequency response of the platform. In these cases,
the image will not be held totally stable. Here
another technique can be employed.
To improve the stability of the image, the
output of SceneLock can be fed into an image warping engine to electronically stabilize
the image by introducing shifts, rotations
and scale changes to cancel out the measured movements.
For example, if SceneLock calculates that
the image is moving to the right, the image
warping will be commanded to shift it to the
left, to cancel out the motion. The measured
motion can be filtered to remove unwanted,
high frequency jitter, but true pans and
zooms can be passed through if required.
This image warping technology from
GE Intelligent Platforms provides a low
latency, highly accurate warped video
stream. The stabilization process is robust
to small independently moving objects
within the scene.
With real-time image stabilization, image quality
is markedly improved in the video stream.
Applied imAge processing
Page 7
VISUALIZATION
Image Fusion.
Improving tactical understanding by fusing multiple
images into a single intelligent picture.
Image Fusion
The number of sensors on military vehicles is
increasing rapidly, leading to a requirement
for intelligent ways to present information to
the operator without information overload,
while reducing the power consumption,
weight and size of systems.
Military and paramilitary imaging systems
can include sensors sensitive to multiple
wavebands including color visible, intensified visible, near infrared, thermal infrared
and tera hertz imagers.
Typically, these systems have a single display
that is only capable of showing data from
one camera at a time, so the operator must
choose which image to concentrate on,
or must cycle through the different sensor
outputs. Image fusion is a technique which
allows us to combine the complementary
information from each sensor into a single,
superior image which can be displayed to
the operator. GE Intelligent Platforms offers
a range of real-time, full-frame video fusion
solutions which aim to maximize scene detail
and contrast in the fused output, producing
superior fused image quality with maximized
information content.
The image fusion algorithm embedded
into our systems is a new approach to
multi-scale fusion which benefits from
Image fusion combines multiple sensor inputs into a single, superior image.
much faster execution times and reduced
memory overheads. This novel algorithm
gives significantly improved results over the
baseline weighted average algorithm while
still performing in real time on live imagery.
This algorithm has been designed especially
for a low latency, real-time implementation
which is flexible enough to fuse images from
a wide range of sensors.
Our built-in warp engine provides rotation, scaling and translation for each video
source to compensate for image distortion
and misalignment between the imagers,
reducing the need for accurate matching of
imagers with a resulting reduction in overall
system cost.
Compared to existing products (either
software-based with limited performance or
using non-ruggedized hardware) our video
fusion solutions offer greater functionality in
a ruggedized and compact form factor.
Page 8
VISUALIZATION
Situational Awareness.
Making a significant contribution to crew safety
by providing total awareness of the local situation.
Situational Awareness Systems
The Challenge
Surveillance of large areas has historically
been achieved by using an array of sensors
connected to a bank of monitors, with separate or multiplexed displays for each video
stream. Such arrangements present the
operator with a confusing array of disparate
video feeds, require a great deal of space
and consume a large amount of power. In a
dynamic, real-time scenario, there is also a
danger of information overload for an operator attempting to consume and interpret
such large volumes of imagery.
The Solution
The GE Intelligent Platforms IPS5000 and
IPS500 systems provide an optimized
solution for local situational awareness by
seamlessly merging the video signals from
multiple sensors into a real-time interactive
360º panoramic image which can be
displayed on a single monitor, saving space
and power. It offers a previously unattainable level of situational awareness to platforms such as armored vehicles, aircraft,
remote unmanned platforms and security
and surveillance systems.
The operator is able to interact with the
panoramic image through a range of interface options including touch screen, joystick
and game-style controller. These devices
facilitate scrolling through the panorama
Our image processing subsystem greatly improves the performance of surveillance assets
and their operators.
and zooming in and out on selected images
through a technique known as ‘virtual
camera.’ Used in this way, the panorama
is very intuitive to use and interpret. The
panorama can also be integrated with a
head-mounted display to form a ‘virtual
control room’ effect.
The IPS5000/IPS500 display layouts can be
adapted to meet specific customer requirements. It has been designed to offer an intuitive user interface with a natural looking,
seamless and undistorted perspective when
seen by the virtual camera
The panorama can be formed with video
images from all types of sensors including
visible, image intensified and thermal and
where appropriate can be fused into composite panoramas
The IPS5000 and IPS500 systems can be
integrated to cue a pan and tilt unit (such as
a commander’s sight) to offer high resolution images of panorama targets.
In addition the IPS5000 and IPS500 systems
can perform ‘data extraction’ on the sensor
images and carry out both short term ‘loop
recording’ and long term ‘archival recording.’
Applied imAge processing
Page 9
Research and
Development.
To keep you on the leading edge, we continuously
explore new image processing technology.
For over two decades, our rolling program
of research and development has delivered
innovative technology in video tracking and
image processing for customers around the
globe. Our goal is to build on this heritage,
so that you stay at the forefront of video image processing technology in the future.
Working to the exacting standards of ISO
9001:2000, our engineers are developing
the next generation of image processing products while incorporating ongoing
product improvements.
Because we are at the cutting edge of new
product design and development, we can
meet the challenges of new applications
head on and give you a range of products
you can rely on.
Spectacular Performance
One of the most spectacular applications of
our image processing technology is in the
autonomous refueling of unmanned aircraft
(UAVs). GE Intelligent Platforms is a partner
in a program which will allow unmanned
aircraft to autonomously refuel in flight. Our
video processing system locates, recognizes
and tracks the refueling connector and
outputs data to the flight control system to
close the loop and effectively steer the UAV
to dock with the refueling aircraft.
Initial demonstrations used a NASA aircraft
to simulate the UAV with the pilots handsoff. Subsequent flights will lead to a fully
autonomous refueling capability.
Our video tracker played a key role in the first
ever, autonomous probe and drogue airborne
refueling demonstration.
Page 10
Fusion
6 Combine inputs from several imager
types to create a single, superior image.
Stabilization
6 Clear, stable images are always less
confusing and more useful to the operator.
Image processing solutions
optimized for SWaP.
GE Intelligent Platforms COTS-based image processing solutions provide
high performance while requiring the minimum space, weight and power.
The SWaP optimized hardware design can be easily reconfigured to meet
restricted space applications such as within armored vehicles.
Page 11
Situational Awareness
6 The IPS5000 provides local situational
awareness by merging video from multiple
cameras into a real-time, interactive,
360º panoramic view.
Detection
6 The system identifies objects within the
video image which meet your target criteria.
Tracking
6 Choose from an array of tracking
capabilities, based on your application needs.
Page 12
GE Intelligent Platforms Contact Information
Americas: 1 800 433 2682 or 1 434 978 5100
Global regional phone numbers are listed by location on our web site at www.ge-ip.com/contact
www.ge-ip.com/imageprocessing
©2010 GE Intelligent Platforms, Inc. All rights reserved.
*Trademark of GE Intelligent Platforms, Inc.
All other brands or names are property of their respective holders.
01.10 10M GFA-1307
Loading...