Award Data

Reducing the False Alarm Rate (FAR) of Automated Target Recognition (ATR) algorithms for Synthetic Aperture Radar (SAR) imagery is crucial for Intelligence, Surveillance, Reconnaissance (ISR) and precision target engagement missions. While modern Deep Learning (DL) ATR networks have demonstrated advanced predictive capabilities and generalization for SAR imagery, they lack spatial awareness, resul ...

The pace of remotely sensed data collection is increasing exponentially while the skilled workforce that is available to analyze and interpret the data is not. Artificial Intelligence (AI) offers a technology pathway to automation which promises to rapidly reduce the gap between data and information products and services. However, AI techniques require massive quantities of labeled data to train m ...

In Phase I, our team of Kitware and UC-Berkeley developed Scale-MAE by adding ground sample distance (GSD) to positional encodings, and produced a multiscale representation that achieves state-of-the art results across image classification, semantic segmentation, and object detection tasks. In Phase II, we will create a remote sensing pretraining toolkit to enable fast and easy experimentation wit ...

Satellite platforms play a critical role in the modern defense and intelligence infrastructure, providing timely, detailed, and readily available imagery to support U.S. national security. An ever present and fundamental requirement for this type of data is automated target detection and recognition, helping to quickly locate and correctly identify targets from vast quantities of image data. Despi ...

In this proposed Phase II effort we will implement a software framework that will reduce the time required to annotate large image/video datasets by a factor of 100x while also reducing the data necessary to train state of the art computer vision models by up to 80%. Our strategy combines several elements and ideas. First, we have developed a sub-tile based a priori theory of how and why CNNs can ...

In Phase I we developed and demonstrated a bi-directional, monolithic, High-Speed, Electro-optic shutter for Range Gated Imaging. While meeting the majority of the key performance parameters for Range-Gated Imaging at high Pulse Repetition Frequency, improvements can be achieved in transmission, contrast and reliability. The Phase Ii activity begins with a upgraded Electro-optic Shutter incorpora ...

Existing THz devices have not yet provided all the imaging functionalities required to fulfill non-destructive imaging and remote sensing applications. Most of the THz imaging and spectroscopy systems utilize single-pixel detectors, which results in a severe trade-off between the measurement time and field-of-view.   Lack of large-format (many-pixel) sensor arrays is one of the major hurdles for ...

Prixarc proposes to develop and commercialize novel terahertz (THz) focal-plane array (FPA) that utilizes 3D microstructures, smart readout integrated circuits, and allows efficient integration with processors that incorporate machine learning to increase the data collection efficiency. We plan to collaborate with University of Miami (UM) and Kansas State University (KSU) for this project. We prop ...

Object detection datasets for overhead imagery are typically generated using bootstrapping methods to reduce annotator effort and cost. These methods iteratively train a detector from a limited set of user-provided and model-predicted labels. Such approaches bias detectors toward the initial object set, limiting their capacity to handle object variations or discover novel objects classes. MONET ov ...

Commercial imaging satellites capture the majority of the planet’s surface every day, generating millions of images. Searching such data for specific objects of interest is extremely difficult to do manually without automated computer image analytics. NGA is seeking a novel scientific formalism and tools for training overhead image detectors on rare and new objects of interest, for which large, ...