Award Data

The objective of this project is to design a software architecture based on densely-connected neural network to perform automatic targetsegmentation and recognition using training datasets of limited size (low-shot). Deep learning architectures have proved to be extremelyeffective at object detection and recognition, but such capability comes at the cost of having large labeled datasets. Such data ...

Signature Research, Inc. (SGR) and Michigan Technological University (MTU) propose a Phase I STTR effort to develop a learning algorithm which exploits the spatio-spectral characteristics inherent within IR imagery and motion imagery.Our archive of modelled and labeled data sets will allow our team to thoroughly capture the variable elements that will drive machine learning performance.The overall ...

This SBIR Phase I project proposes a probabilistic approach to determine a vehicles location using onboard video sensors and foundationalmap data. The system does not rely on only one type of information source, instead it combines proposals from a variety of locationestimators to find a vehicles location in GPS-denied environments.The system takes advantage of recent advancements in computer visi ...

Deep Learning Analytics (DLA) will develop the Deep-False Alarm Suppression Technique (D-FAST) algorithm that uses state of the art and

Low-temperature MBE grown GaAs (LT-GaAs) contains a high concentration of excess As which gives rise to ultra-fast carrier-trapping time and excellent radiation hardness. In as-grown layers most of this excess As is in the form of As(Ga) antisite defects, of which only ~1% are ionized. Thermal annealing upon overgrowth with a device structure or during device processing results in a decrease of ...