Image Labeling for Sparse and Rare Targets in Satellite Imagery for Computer Vision Applications

The Department of Defense (DOD) is seeking proposals for the Small Business Innovation Research (SBIR) program focused on "Image Labeling for Sparse and Rare Targets in Satellite Imagery for Computer Vision Applications." The objective is to refine existing image labeling methodologies to enhance the efficiency and cost-effectiveness of producing datasets that contain rare and challenging objects of interest (OOIs) in overhead satellite imagery. This initiative is crucial for improving computer vision detector algorithms, particularly in scenarios where OOIs are scarce, lack distinct features, or are subject to occlusion and natural obfuscation. Interested parties should note that the work will be conducted on classified data at the TS/SCI level, with Phase I proposals due by May 21, 2025, following the opening of the solicitation on April 23, 2025. For more information, please visit the DOD SBIR website at https://www.dodsbirsttr.mil/topics-app/.