NASA STTR 2010 Solicitation

FORM B - PROPOSAL SUMMARY


PROPOSAL NUMBER: 10-1 T5.01-9939
RESEARCH SUBTOPIC TITLE: Technologies for In Situ Compositional Analysis and Mapping
PROPOSAL TITLE: Real-Time Smart Tools for Processing Spectroscopy Data

SMALL BUSINESS CONCERN (SBC): RESEARCH INSTITUTION (RI):
NAME: Signal Processing, Inc. NAME: U. Tennessee
STREET: 13619 Valley Oak Circle STREET: 1508 Middle Drive
CITY: Rockville CITY: Knoxville
STATE/ZIP: MD  20850 - 3563 STATE/ZIP: TN  37996 - 2100
PHONE: (301) 315-2322 PHONE: (865) 974-8527

PRINCIPAL INVESTIGATOR/PROJECT MANAGER (Name, E-mail, Mail Address, City/State/Zip, Phone)
Chiman Kwan
chiman.kwan@signalpro.net
13619 Valley Oak Circle
Rockville, MD 20850 - 3563
(301) 315-2322

Estimated Technology Readiness Level (TRL) at beginning and end of contract:
Begin: 2
End: 4

TECHNICAL ABSTRACT (Limit 2000 characters, approximately 200 words)
We propose novel and real-time smart software tools to process spectroscopy data. Material abundance or compositional maps will be generated for rover guidance, sample selection, and other scientific missions. First, we propose a novel anomaly detector called clustered kernel Reed-Xiaoli (CKRX) algorithm. This tool was developed by us, is fast, and can achieve very high anomaly detection rate in hyperspectral images from the Air Force. This is important in planetary missions because we may need to look for some anomalous regions in a scene. Second, if target material signatures are available, then we propose a fast matched signature identification algorithm called Adaptive Subspace Detector (ASD). We compared ASD with several other tools and found that ASD outperformed other methods. Third, if target material signatures are not available, then we propose a new technique called minimum volume constrained non-negative matrix factorization (MVCNMF) to perform unsupervised material identification. In a recent comparative study by using hyperspectral images from the Air Force, the MVCNMF performed better than some conventional unsupervised methods. Fourth, the above tools can be implemented in a parallel processing architecture, in which the computations are distributed to multiple cores. We have applied it to speech processing and genomic processing recently. Real-time performance is achievable.

POTENTIAL NASA COMMERCIAL APPLICATIONS (Limit 1500 characters, approximately 150 words)
Our proposed algorithm can exactly meet the NASA's mission needs, including rover guidance, sample selection, and other scientific missions. We can handle different scenarios such as anomaly detection, supervised material identification, and unsupervised material identification.

POTENTIAL NON-NASA COMMERCIAL APPLICATIONS (Limit 1500 characters, approximately 150 words)
We expect to produce real-time tools containing the above mentioned algorithms for hyperspectral and multispectral image processing. The tools can be useful for military surveillance and reconnaissance, and civilian applications (vegetation monitoring). The market size is estimated to be 20 million dollars over the next decade.

TECHNOLOGY TAXONOMY MAPPING (NASA's technology taxonomy has been developed by the SBIR-STTR program to disseminate awareness of proposed and awarded R/R&D in the agency. It is a listing of over 100 technologies, sorted into broad categories, of interest to NASA.)
Algorithms/Control Software & Systems (see also Autonomous Systems)
Data Processing
Image Processing
Multispectral/Hyperspectral
Software Tools (Analysis, Design)


Form Generated on 09-03-10 15:17