The Center for Optical Signature Recognition strives to conceive and demonstrate useful leading-edge technologies involving optical sensing, discrimination and recognition. These technologies include sensor hardware and signal/pattern processing algorithms, matched to deduce useful and reliable information from various sensing situations. We develop methods essential to the conception, design, assessment, and improvement of civil and military geophysical remote sensing, surveillance, seeking, hiding, and advertising systems employing an optical (electro-optical/infrared) component. The optical spectral regime in which we operate spans the ultraviolet, the visible, and infrared regions (0.1 - 20 micrometers). We are expert in passive sensing, exploiting the best from the full range of observable features which include the spatiotemporal radiometric, spectral (colorimetric), and polarimetric signatures (appearance). Our R&D continually pursues both application-specific and general solutions to the following problems:
- Which observables (e.g., spectral, spatial, temporal, polarimetric) "best" differentiate targets/constituents of interest from non-targets?
- What physical sensing concepts are appropriate and feasible for transducing the maximum "information"?
- How can algorithmic techniques robustly discriminate targets and infer desired "state" information?
- What are the designs and performance bounds for systems which maximize or minimize discriminability?
We serve both government and private industry clients, including the R&D labs of the military services, NASA, NOAA, and prime aerospace contractors. We also pursue collaboration with academia; for instance we are an industrial affiliate of Northeastern's NSF-sponsored Center for Subsurface Sensing and Imaging Systems (CenSSIS).
Research & Technology Areas
- Detection, tracking, recognition, and surveillance of ground, sea, air and space vehicles
- Geophysical and climate-change research via remote sensing of macro- and microphysical properties
- Non-invasive medical diagnostics/screening
- Object appearance, visibility, conspicuity (e.g., camouflage) engineering and optimization
- Pattern Recognition/Detection & Estimation Theory
- Computational Vision & Perception
Optical Sensing Systems Aspects
- Electro-optical systems technology
- Radiative transfer/atmospheric propagation models
- Optical properties of materials
- End-to-end physics-based simulations of sensor-object|scene encounters.
- Optical Remote Sensor Lab, supporting the assembly, alignment, calibration and testing of optical sensors. Available LN2 (liquid nitrogen), roughing and ion vacuum pumps are particularly useful in working with infrared cryogenic sensors. Illumination sources include mercury arc lamps, blackbodies, HeNe lasers, a novel xenon (9um) laser, argon ion lasers, and HeCd lasers.
- Thermal infrared imager, employing microbolometer uncooled focal plane array (UFPA), 3-14 micrometers
- Field-grade Fourier-transform IR spectrometer, with HgCdTe and InSb detectors, 10-inch telescope, 2-14 micrometers, 1 cm-1
- Digital data and video image acquisition computer workstations
- Heterogeneous software development and algorithm prototyping environments
- Physics-based target & scene simulation codes
- Computer-based design tools for optical appearance/ conspicuity engineering of objects.