Research and Development Projects
Martingale Research has been involved since its inception in the development of advanced technologies, under the sponsorship of a variety of government agencies. We have directed our research and development efforts since our company was established towards the development of intelligent computing solutions. Our research and development efforts have advanced the state of the art in pattern recognition, signal and image processing, statistical modeling, and data analysis.
This project developed a new approach to database analysis called Constrained Categorical Regression (CCR). CCR is a modeling methodology that integrates prior knowledge as logical causal relationships directly into the analysis of categorical dependent-variable data. It makes use of both neural network and advanced statistical technologies to provide advanced capabilities for modeling and exploiting data relationships.
Under the sponsorship of the National Science Foundation, Martingale Research developed an intelligent computing technology for detecting, classifying, and predicting fatigue damage in metals and composite structures. Based on Acoustic Emissions processing, it is suited for real-time, embedded applications such as smart structures and will also be available for use in smart nondestructive testing instruments.
The Parametric Avalanche Stochastic Filter (PASF) was developed and refined over a period of many years as an advanced technology for detection, tracking, and control of dynamical systems. This development has resulted in a neural network-based processing architecture with applications in many areas including speech recognition, signal and image processing, and control systems.
Martingale Research has developed algorithms for detection and estimation of motion in imagery. The technology is applicable in a variety of applications, including video compression (e.g. MPEG), surveillance, target tracking, and image understanding.