SBIR/STTR Research Award

MsObjDET: AI-based Multi-Spectral Passive Object Detection using Spatiotemporal Transformer and YOLO with Dynamic Tracking Integration

Award identifier 296afa3bbeb5dccc82dd6e8be2603dc43e6e5ac548b1a12ff7558717e45d05eaAwarding agency Department of Defense · Navy

Award amount

$139,999

Program and phase

SBIR · Phase I

Award year

2025

Explore This Research Award in CaptureIQ

Research the technology, evaluate the recipient and agency, identify related opportunities, and explore potential follow-on work with CaptureIQ.

Award details

Awarding agency
Department of Defense
Branch
Navy
Recipient company
INFOBEYOND TECHNOLOGY LLC
Proposal award date
2025-06-11
Contract end date
2025-12-08
UEI
JMXHCAK12LZ7
Solicitation number
25.1
Topic code
N251-052

Research abstract

The complexity of multi-observer and multi-spectral passive object detection in maritime environments limits the Navyís ability to track, classify, and identify surface and aerial targets with high confidence. Traditional methods suffer from sensor misalignment, environmental interference, and suboptimal spectral fusion, leading to detection inaccuracies and delayed situational awareness. In this project, InfoBeyond proposes MsObjDET, an AI-enhanced passive detection system that fuses multi-spectral sensor data from shipborne EO/IR arrays, UAV-mounted platforms, fixed maritime stations, and radar systems to enhance real-time object detection and tracking. At first, MsObjDET includes a Multi-Spectral Spatiotemporal Transformer to deeply extract extracts spectral and spatial features across signals. ÝThe Transformer employs a Spectral Competitive Mechanism to dynamically evaluate the reliability and relevance of spectral features under operational noises. ÝMeanwhile, a Spatiotemporal Attention is designed to capture temporal dynamics and spatial correlations, enabling the alignment of multi-spectral data streams across time and space. ÝSecondly, the YOLO Detection Head is proposed to identify and localize targets with high precision, addressing challenges such as small object detection, multi-target interference, and dynamic background noise. Further, a Dynamic Tracking & Integration module is proposed to learn a unified and high-confidence tracking Common Operational Picture (COP), addressing challenges such as trajectory fragmentation, temporal inconsistencies, and multi-target interactions. Superior to existing system, MsObjDET corrects spectral inconsistencies, compensates for platform motion over time, and integrates confidence-aware detections to provide precise object classification and tracking across Aegis-integrated systems. Designed for seamless deployment within the Aegis Combat System (ACS), including AN/SPY-1 radar and passive radar arrays, MsObjDET enables persistent surveillance of high-threat regions while mitigating sensor limitations.Ý

Technology keywords

Not reported

Other awards by this recipient

Verified by the same UEI. These are other public SBIR/STTR award pages, not a merged company record.

Awards with the same topic code

Verified by an exact topic code match. This is not a text-similarity match.

Research funding history

44 SBIR/STTR records share this UEI (2010–2025). Reported award amounts total $13,013,950. Reported amounts are not federal obligations.

Phases: Phase I, Phase II

Agencies: Department of Defense, Department of Energy, Department of Commerce, Department of Transportation

Agency intelligence

Department of Defense

Back to SBIR/STTR Research Awards

Don't take our word for it. Ask AI about RFPCapture.

See what leading AI assistants have to say about RFPCapture and CaptureIQ.

Gemini opens without a prefilled question. The question is copied to your clipboard when the button works. You can also copy it here:

What is RFPCapture.com and its CaptureIQ platform from Smart Logic AI? How does it help federal contractors discover opportunities, evaluate bid/no-bid decisions, research government contracts, and prepare proposals? How does it compare with other federal capture platforms? Use independent sources where available.