Systems & Technology Research LLC — Department of Defense SBIR Phase I: NGA201-004
Systems & Technology Research LLC — SBIR Phase I award from Department of Defense.
Phase I SBIR feasibility signal
- Phase I awards fund proof-of-concept work. For capture teams, they mark early interest from Department of Defense in a technical approach.
- Watch for Phase II follow-ons from the same firm/topic family — that conversion path is where budgets and transition pressure rise.
- Obligated amount $95,640. Cross-check similar awards in the same agency and technology tags for going-rate context.
- Topic code NGA201-004 links this award to a solicitation family — search the same topic stem for incumbents and recompete timing.
- Amount
- $95,640
- Agency
- Department of Defense · National Geospatial-Intelligence Agency
- Program / Phase
- SBIR · Phase I
- Topic
- NGA201-004
- Solicitation
- 00.1
- NAICS
- —
- Place of performance
- MA
- Period
- 2021-01-29 → 2021-10-27
Description
New satellite constellations will soon be deployed with locational revisit rates of 40 captures/day resulting in an image acquisition volume that quickly exceeds capacity for manual image analysis. This inspires an automated solution for analyzing spatio-temporal imagery that can augment analyst workflow and enhance geospatial investigation of locations over time. Building on System & Technology Research’s (STR’s) previously developed spatio-temporal software and using STR’s access to over 20 petabytes of commercial satellite imagery, we propose the Spatio-Temporal Analysis & Recognition System (STARS) solution. STARS incorporates integrated uncertainty characterization and temporal evidence aggregation for reliable object detection and disambiguation, a mature data conditioning pipeline, robust few-shot learning approach, and extensive object detection experience with diverse satellite datasets. This solution is designed to identify, detect, and disambiguate rare objects across spatio-temporal sequences of images. Our proposed system leverages feature enhancement and reweighting to learn optimal features for few-shot object detection. Detections can be leveraged by our proposed Bayesian non-parametric probability estimation method to provide meaningful and quantifiable confidence estimates and be tolerant to variations in collection resolution, orientation, and scene illumination.