Data Exploitation Group - Mission

Michigan Aerospace Corporation’s Data Exploitation Group applies data mining, pattern recognition, machine learning and image processing technologies to develop products for the defense and civil space communities, as well as the private sector. The staff of this group has over three decades of experience in delivering pattern recognition solutions to the United States government and private industries, including chemical companies, pharmaceutical, biotechnology, and automotive. We have the combination of technologies and application experience in data exploitation software to understand our customers' problems and deliver quality software solutions.

Novelty and Anomaly Detection

MAC's Data Exploitation Group provides customizable software and services for anomaly and novelty detection. Products for discovering novel events and detecting anomalous behavior are quickly becoming indispensable for the proper operation and maintenance of the complex systems employed by modern industry, medical providers and the military. Factories, health monitors, aircraft and other vehicles regularly produce hundreds or thousands of channels of telemetry in real time, which must be monitored for possible indications of failure. These data resources present an enormous opportunity for software tools that can detect events of interest. These tools can be used to discover useful new insights into operations, diagnose faults post mortem to determine the cause, and predict failures before they cause costly damage or injury.

Our group has developed powerful pattern recognition technology, based on Ensembles of Decision Trees (EDTs), which drives our tools for novelty and anomaly detection. EDT-based technology has many advantages over existing techniques, including the scalability to handle very high dimensional data, automatic drill-down facilities to explain why an event is categorized as novel or anomalous, and extremely rapid evaluation. The approach is also amenable to deployment on FPGA devices, Graphical Processing Units, parallel processors and other accelerated platforms.

Pattern recognition technology developed by Michigan Aerospace Corporation

Pattern Recognition / Machine Learning / Data Mining

MAC's staff have intimate knowledge and practical experience using the approaches seen in the following table:

Neural Networks	Decision Trees and Ensembles	Support Vector Machines
K Nearest Neighbors	Clustering Techniques	Bayesian Learning
Evolutionary Algorithms	Hidden Markov Models	Self Organizing Maps (SOMs)

Computer Vision and Image Processing

The Data Exploitation Group’s scientists have accumulated a wealth of expertise in arenas related to computer vision including image processing, visualization, machine learning, optimization, and pattern recognition.

Our scientists have expertise in many facets of image analysis including:

Mathematical Morphology
Statistical Image Processing
Fourier and Wavelet Analysis

Using these low-level paradigms our researchers have developed and implemented algorithms in previous projects crucial for computer vision tasks such as:

Preprocessing, Calibration, Normalization, and Enhancement
Feature Extraction
Measuring of Image Attributes (Texture, Shape, Color, etc.)
Registration and Mosaicking
Segmentation and Thresholding
Object Detection, Classification and Recognition
Target Tracking

These algorithms have been developed for a broad array of imaging modalities from many arenas including:

3D laser range and reflectance
Multispectral imagery (Thematic Mapper [TM], LandSAT, and Spot)
Synthetic Aperture Radar (SAR) imagery
Stereo pair imagery

Light microscopy
Electron micrographs (EM, STEM)
Confocal microscopy
Computed Tomography (CT)
Magnetic Resonance Imagery (MRI)
Scanned Documents