Description: OBJECTIVE: Develop a wide area standoff hyperspectral-imaging sensor for chemical and biological early warning based on compressed sensing. DESCRIPTION: Current hyperspectral imaging (HSI) sensors in use for standoff chemical/biological sensing generate huge amounts of data. To save space or to reduce bandwidth for transmission the data is often compressed. Lossy compression algorithms for hyperspectral imaging have had broad success. Using state-of-the art data compression algorithms researchers have discovered that much of the data from existing HSI sensors can be thrown away with little or no loss in standoff detection capabilities. This fact has led researchers to explore methods that limit data acquisition to data that is not rejected by standard compression algorithms. The goal of this effort is to examine compressed sensing as a method for reducing the size, weight, power, cost, and bandwidth of current HSI systems, while still providing high sensitivity and the capability of wide area early warning of a chemical or biological attack. Compressed sensing has evolved as a method to simplify HSI sensors and improve standoff chemical/biological sensing. Compressed sensing offers simultaneous compression and sensing processes, based on the existence of a sparse representation of a signal within a set of projected measurements. Compressed sensing looks at possibility of acquiring only the data that is needed for detection and avoiding the acquisition of data that is redundant or not applicable to the problem at hand. Compressed sensing also looks at methods to extract the maximum amount of information from reduced data sets. Most work to date has concentrated on systems with plentiful signal, low noise, and sparse information. HSI sensors for standoff chemical/biological sensing, however, often work in spectral regimes of low signal-to-noise ratio, and therefore may require unique approach to signal compression in both hardware and data processing to maintain signal to noise ratio during the compression process. Hyperspectral imaging sensors currently in use for wide-area standoff detection of chemical and biological agents are required to utilize large focal-plane-arrays to achieve necessary spatial coverage and spatial resolution for wide area chemical/biological detection. They are also required to interrogate a large number of spectral bands in order to differentiate between target compounds and the background. In addition to these very stringent hyperspectral-imaging requirements, a chemical/biological standoff sensor needs to be small, lightweight, and inexpensive. Current standoff chemical/biological sensors operate in the 8 to 12 m wavelength range to access important CB and toxic industrial material signatures. The use of coded apertures and spatial light modulators may provide a method of using compressed sensing in a manner that eliminates the need for an infrared focal-plane-array in a wide-area standoff chemical/biological sensor. Coded apertures may provide a method of extracting spatial and spectral information from a scene using a single-pixel infrared detector. The use of a single pixel detector in the place of a focal-plane-array has the potential to reduce size, weight, power, and cost of a wide area standoff chemical/biological sensor without sacrificing performance parameters. The key factor for long-wave CB monitoring is to develop compression algorithms that maintain detection sensitivity and signal to noise performance, while minimizing the data throughput requirements. The use of adaptive coded apertures may also provide a method of dynamic foveation, where a portion of the dynamic scene can be queried in greater detail if an anomaly is detected. Additionally, signal compression in the spectral domain can be achieved by aperture coding algorithms that employ spectral detection filters to discriminate the analyte signature relative to that of the background with optimum detection sensitivity and false alarm rate. PHASE I: Design a single-pixel hyperspectral imaging sensor for wide area standoff detection of chemical and biological agents based on compressed sensing. The spectral region of the sensor should be chosen to interrogate spectral signatures of chemical plumes. Traditionally the 8 to 12 m region of the electromagnetic spectrum has been used for standoff chemical detection. Examine the use of coded apertures to produce a single pixel sensor with detection and discrimination capabilities comparable to existing HSI chemical/biological sensors. The goal is to passively detect small chemical plumes (25 meters or smaller) of a chemical agent such as sarin at relevant concentrations (a few ppmv) at a distance of 5 kilometers or more under ambient conditions. The system should be designed to reduce size, weight, and power compared to traditional HSI systems while maintaining similar detection limits (equivalent to monitoring differential radiance signals of the order of 1 microflick or less on a 1 second time scale). Examine methods of dynamic foveation within a hyperspectral image using compressed sensing. Examine the use of dynamic coded apertures to query a portion of the dynamic scene based on the detection of an anomaly or other alarm. Examine the use of spectral compression algorithms based on coded apertures to discriminate the chemical plume against the scene background. PHASE II: Construct a standoff hyperspectral imaging sensor designed for the detection of chemical plumes. Utilize the best methods and technologies for reducing the size of current HSI systems while maintaining required sensitivities. Test and characterize the new HSI sensor. Based on the tests, update the design of the new standoff chemical imaging sensor. Deliver the prototype sensor to the government. PHASE III DUAL USE APPLICATIONS: Further research and development during Phase III efforts will be directed towards refining a final deployable design, incorporating design modifications based on results from tests conducted during Phase II, and improving engineering/form-factors, equipment hardening, and manufacturability designs to meet U.S. Army CONOPS and end-user requirements. There are many environmental applications for a small chemical standoff sensor. A rugged, sensitive and flexible chemical detector will benefit the manufacturing community by providing finely tuned monitoring of chemical processes. Also first responders such as Civilian Support Teams and Fire Departments have a critical need for a rugged, relatively inexpensive but versatile and rugged sensor that can be transported to the field to test for possible contamination by CW agents and other toxic chemicals. REFERENCES: 1. Bogdan R. Cosofret, Shin Chang, Michael L. Finson, Christopher M. Gittins, Tracy E. Janov, Daisei Konno, William J. Marinelli, Mark J. Levreault, and Rex K. Miyashiro,"AIRIS standoff multispectral sensor", Proceedings of the SPIE, volume 7304, pages 73040Y (2009). 2. Vincent Farley, Charles Belzile, Martin Chamberland, Jean-Francois Legault, and Karl R. Schwantes,"Development and testing of a hyperspectral imaging instrument for field spectroscopy", Proceedings of SPIE, volume 5546, pages 29-36 (2004). 3. Marco F. Duarte, Mark A. Davenport, Dharmpal Takhar, Jason N. Laska, Ting Sun, Kevin F. Kelly, and Richard G. Baraniuk,"Single-Pixel Imaging via Compressive Sampling", IEEE Signal Processing Magazine, volume 25, issue 2, pages 83-91 (2008). 4. Jianwei Ma,"Single-Pixel Remote Sensing", IEEE Geoscience and Remote Sensing Letters, volume 6, issue 2, pages 199-203 (2009). 5. D.J. Brady and M.E. Gehm,"Compressive imaging spectrometers using coded apertures,"Proceedings of the SPIE, volume 6246, pages 62460A (2006). 6. D. L. Donoho,"Compressed sensing,"IEEE Transactions on Information Theory, volume 52, pages 1289-1306 (2006). 7. E. J. Candes and T. Tao,"Near-optimal signal recovery from random projections: Universal encoding strategies?", IEEE Transactions on Information Theory, volume 52, page 5406-5425 (2006). 8. Jianwei Ma and F.-X. Le Dimet,"Deblurring From Highly Incomplete Measurements for Remote Sensing", IEEE Transactions on Geoscience and Remote Sensing, volume 47, issue 3, pages 792-802 (2009). 9. Dharmpal Takhar, Jason N. Laska, Michael B. Wakin, Marco F. Duarte, Dror Baron, Shriram Sarvotham, Kevin F. Kelly, and Richard G. Baraniuk,"A New Compressive Imaging Camera Architecture using Optical-Domain Compression", Proceedings of the SPIE, volume 6065, pages 43-52 (2006). 10. R. Larcom and T. R. Coffman,"Foveated image formation through compressive sensing", 2010 IEEE Southwest Symposium on Image Analysis & Interpretation (SSIAI), pages 145-148 (2010). 11. Vujkovic-Cvijin, P., Goldstein, N., Fox, M.J., Higbee, S.D., Becker L.S., and Ooi, T.K."Adaptive Spectral Imager for Space-Based Sensing,"Proceeding of the SPIE Volume 6206, paper 6206-33 (2006). 12. Goldstein, N., P. Vukovic-Cvijin, M. Fox, S. Adler-Golden, J. Cline, B. Gregor, J. Lee, A. C. Samuels, S. D. Higbee, Latika S. Becker, and Teng Ooi,"Programmable Adaptive Spectral Images for Mission-Specific Application in Chemical/Biological Sensing,"International Journal of High Speed Electronics and Systems, volume 17, issue 4, pages 749-760 (2007). 13. Goldstein, N., P. Vujkovic-Cvijin, M. Fox, B. Gregor, J. Lee, J. Cline, and S. Adler-Golden"DMD-based adaptive spectral imagers for hyperspectral imagery and direct detection of spectral signatures,"Proceedings of the SPIE, volume 7210, pages 721008/1-721008/10 (2009).