It is critical in military applications to be able to extract features in imagery that may be of interest to
the viewer at any time of the day or night. Infrared (IR) imagery is ideally suited for producing
these types of images. However, even under the best of circumstances, the traditional approach of
applying a global automatic gain control (AGC) to the digital image may not provide the user with
local area details that may be of interest. Processing the imagery locally can enhance additional
features and characteristics in the image which provide the viewer with an improved understanding
of the scene being observed. This paper describes a multi-resolution pyramid approach for
decomposing an image, enhancing its contrast by remapping the histograms to desired pdfs, filtering
them and recombining them to create an output image with much more visible detail than the input
image. The technique improves the local area image contrast in light and dark areas providing the
warfighter with significantly improved situational awareness.
Super resolution reconstruction (SRR) improves resolution by increasing the effective sampling frequency. Target acquisition range increases but the amount of increase depends upon the relationship between the optical blur diameter and the detector size. Range improvement of up to 52% is possible.
Modern systems digitize the scene into 12 or more bits but the display typically presents only 8 bits. Gray scale compression forces scene detail to fall into a gray level and thereby "disappear." Local area processing (LAP) readjusts the gray scale so that scene detail becomes discernible. Without LAP the target signature is small compared to the global scene dynamic range and this results in poor range performance. With LAP, the target contrast is large compared to the local background. The combination of SRR and LAP significantly increases range performance.
It is critical in surveillance applications to be able to extract features in imagery that may be of interest to the viewer at
any time of the day or night. Infrared (IR) imagery is ideally suited for producing these types of images. However, even
this imagery is not always optimal. Processing the imagery with a local area image operator can enhance additional
features and characteristics in the image that provide the viewer with an improved understanding of the scene being
observed. This paper discusses the development of two algorithms for image enhancement for infrared imagery using
local area processing. The enhancement algorithm extends theory previously developed for medical applications.
Algorithm differences addressed include application to IR imagery and to a panning camera rather than still imagery. It
also discusses the obstacles encountered and overcome for insertion of this algorithm into a 10" gimbaled midwave
infrared imaging system for a variety of real-time processing applications. This technology is directly applicable to
driver's vision enhancement systems as well as other night visions systems such as night vision goggles.
Through the trade-off temporal information, a significant increase in spatial resolution is obtainable. This improvement
is quantifiable by using Airy's disc analysis against camera sensor pitch. Integrate the use of Airy's disc to quantify the
image improvement in resolvability and ultimately system range. It this comparison that sets the ground works for
realistic expectation. Our SR system is a natural tracker of moving vehicles with the addition of improved target
resolvability. Super Resolution can capitalize on camera platforms instability. A by product of SR is digitally stabilize
imagery to a fraction of a sub-pixel. Investigation in the sub-pixel remapping has lead to the developed of improved
super resolve images. Another, approach has lead to the development of a window management scheme for further
improvement. The cleaner, from a noise and structural point-of-view, the composite SR image is the more favorable it is
to high-sharpening. Mapping into a transform space greatly reduces the correlation complexity which makes it easier to
realize the complete algorithm into hardware. We have implemented this system into a real-time architecture. The
hardware configuration is composed of an FPGA and supporting processor.
Hierarchical Target Model Analysis (HTMA) is an automatic pattern matching process for categorizing tactical targets. Stored target model information is re-projected into the image space using the sensor camera model state vector. The analysis is carried out in image gradient angle space for greater flexibility and reduced processing. Re-sampling the gradient angle space allows the classification process to work at a wider variety of target ranges. The target model database is built from an assortment of both target operating and background environmental conditions. Incremental classification is possible by applying the matching strategy at increasing target resolution levels that are either self or range closure induced. The first application of this process has been on thermal imagery. It can easily be extended to other image domains.
The first step in an automatic image target acquisition system is determining the location of candidate objects. Screening for targets must also be done within a tactical scenario timeframe. The screening process must only require a portion of the processing workload since other algorithms must execute in the same time frame. The detection of these candidate objects is allocated to two functions within the same algorithm. The first is a pre-screener and other is a clutter rejection component that will categorize the object nomination into target or non-target classes. This paper describes a screener that meets the necessary requirements for tactical operations. It uses the magnitude and direction of the image gradient. Locations are nominated by looking at local neighborhoods in this gradient space. Regions of interest are then selected and various features are extracted. These features are selected both for their information content and their ease of calculation. Using a Bayes approach, target candidates are selected as plausible targets of interest.
In order to demonstrate lock-on-after-launch (LOAL) capability, imaging infrared missile systems of the future require the ability to autonomously identify and track targets of interest. A robust algorithm architecture must have the flexibility to accommodate the fluid system requirements that drive its design. This paper describes a method to autonomously acquire and track an extended range target through its entire flight scenario. A proven automatic target recognition (ATR) approach is used to detect and identify targets of interest, separating them into non-targets and clutter. The methodology uses a down selection strategy to nominate targets for terminal track. Once nominated, a Weighted Edge Tracker (WET) is employed. The tracker relies upon correlations of appropriately weighted edge directions from frame-to-frame images and reference templates. This combination of automatic target recognition and terminal tracking provides a sophisticated yet simple approach to many challenging long range tracking problems.