Ctrl-P to Print
Multiple Optical Non-Redundant Aperture Generalized Sensors (MONTAGE)
Program Manager: Dr. Dennis Healy
Overview
Emerging DoD and commercial scenarios envision large-scale deployment, coordination and monitoring of ubiquitous imaging systems to keep an eye on complex environments. Stringent challenges in cost, platform related volume, weight, form factor constraints, and the effective exploitation of the resulting large volume of imagery motivate a principled re-examination of the structure, function, and roles of traditional visible and IR cameras.
Imaging has traditionally been defined as a process of measuring object attributes (potentially time varying) as a function of spatial coordinates. The most familiar visible-light imaging technology is provided by the venerable camera, whose lens collects light generated or reflected by objects and maps it onto a light-sensitive medium (film or electronic sensors), thereby capturing a representation of the spatial distribution of the objects within the constraints of a particular 3D to 2D mapping. Over the past 50 years a wide variety of novel imaging systems have been developed for operation in different regions of the electromagnetic spectrum (from radio waves to X-rays) in order to serve a broad spectrum of applications ranging from astronomy, microscopy, medicine, and defense. Since the constraints and requirements on these systems are dramatically different, it follows that the design principles and structures of these systems also vary significantly, and in particular, often differ substantively in format and function from the conventional visible imaging systems described above.
Recent advances in technologies for optical wavefront manipulation, optical detection and digital post-processing have opened new possibilities for imaging systems in the visible and IR regimes, suggesting the development of imagers which differ dramatically in form, fit, and function from time-honored camera designs. Extensive cross fertilization of mathematical formulations and system architectures from different imaging modalities referred to earlier is expected to result in quantum leaps in the performance of more familiar imaging systems.
The MONTAGE program seeks to develop and demonstrate truly revolutionary imaging systems obtained by intelligent integration of the advancing capabilities of the individual optical, detection, and processing subsystems. This integration will exploit recent advances in system optimization methods, which provide an emerging capability for co-design and joint optimization of the optical, detection, and processing aspects of imagers. This new approach stands in marked contrast to traditional design practice involving separate optimization of each subsystem within the confines of its conventional function. It is anticipated that the MONTAGE program will consider imaging system designs in which the traditional roles of subsystems are modified significantly and may even merge, enabling dramatic new formats and capabilities.
A particular objective of this program is the systematic exploration of the trade-offs among analog optical processing as realized by pre-detection optics, on-chip processing within sensor arrays, and post-detection digital processing. The impact of these trade-offs on systems-level performance is a primary concern. In particular, the output from DoD imaging systems is often used in image exploitation tasks like target detection, identification, and tracking. Consequently, optimization of systems with respect to performance relevant to image exploitation tasks is one of the goals of MONTAGE.
Program Goals and Specific Technical Advances:
Current imaging systems for visible and thermal wavelengths employ standard lenses to form an image in the focal plane, where the light intensity distribution is converted into electrical signals by a high resolution detector array. The resultant signals may then be post-processed by a digital subsystem in order to perform calibration, noise suppression or other higher-level operations (such as segmentation and recognition). The performance of the system is often quantified in terms of the total field of view, the angular resolution (ability to distinguish two closely spaced point objects) and signal-to-noise ratio. These performance parameters are impacted in a complex, interrelated manner by the choice of aperture and focal length of the optical subsystem, the pixel size and total number of pixels in the detector array, and finally by the algorithms applied to detector output in the digital post-processing subsystem.
Deployment of high performance imaging systems on a wide variety of platforms (dismounted soldiers, ground vehicles, UAVs) is impacted by the system form factor, which is dictated largely by the aspect ratio of width-to-depth. In conventional imaging systems this is primarily determined by the aperture (width) and focal length (depth) of the imaging optics. Since the focal length is typically greater than the aperture diameter, these systems are "deep." Beyond the obvious impact of this fact on the geometric "fit" of the imager with the platform, as there are often important implications for many other properties of the imaging system, such as its weight and angular moment of inertia.
It should be noted that the angular resolution is primarily limited by the detector pixel size. The diffraction-limited angular resolution of the lens is approximately one order of magnitude better than the resolution achieved by the detector. The depth of these cameras (excluding housing and other packaging) are typically greater than 50 mm. The larger pixel size and smaller number of elements in the LWIR detector array (as compared to the visible detector array) limits both the angular resolution as well as the field of view of these thermal imagers. It is possible to manipulate the angular resolution and the field of view by adjusting the focal length, but this has a direct impact on the system depth.
Currently there is a significant potential for realizing imaging systems differing dramatically in form, fit, and function from the conventional cameras characterized above. Radical transformation of the camera is unlikely to result solely from continued incremental improvement of traditional components and subsystems of imagers through application of ongoing advances in technologies for optical wavefront manipulation, optical detection, and digital post-processing. To transcend this, one should consider truly revolutionary system designs based on intelligent and novel combinations of the advances within these component technologies. Such integrated designs are enabled by recent developments in system optimization methods, which provide an emerging capability for jointly optimizing the individual optical, detection and processing subsystems. This new approach stands in marked contrast to traditional design practice involving separate optimization of each subsystem within the confines of its conventional function. It is now possible to design imaging systems in which the traditional roles of subsystems are modified significantly and may even merge, enabling dramatic new formats and capabilities.
The MONTAGE program seeks to realize radical improvements in the performance and format of imaging systems in order to transform the use of these systems in DoD applications. The program will develop and demonstrate innovative methods for creating and prototyping imaging systems with desirable properties. Specific illustrations of these new capabilities will be provided by the design and test of representative demonstration systems, with evaluation of format and performance in comparison to appropriate baseline systems.