Chemistry and Material Science

Our work in this technology focuses on the understanding and development of chemical and mechanical properties of substances that further our advancement in science and technology.

Research interests in this area span form the development of advanced multifunctional composites and coatings to the formation of inert pseudo-scent aids for training explosive detection canines.

>>>

Test Technology

We offer automated testing solutions for both board and system testing in the government and private sectors.

The inefficiency of laborious manual testing spurred us to expand our technology to include the benefits of automation with the increased availability of emerging technology to provide fast, accurate, and inexpensive identification of failed circuits, either in electronic circuit boards or on the chips themselves.

We design and develop testing strategies and solutions that efficiently exercise the functionality of the unit under test. We achieve a high degree of accuracy and fault diagnostic to the component level. One main focus has been in the design and development of Computer Assisted High-Speed Stimulus Pattern and Timing Set Generation. Our research interests include also unconventional Testing Methods.
>>>

GIS, Imaging, and Visualization

Our imaging research focuses on analysis of images for detection of anomalies such as foreign objects and defects in manufactured materials, as well as features of interest related to pattern recognition problems.

Many techniques exist for detection of objects in images through utilization of multiresolution decomposition in conjunction with neural networks. However, the common joint decomposition/training approach does not address the primary problem of requiring a large representative data set for training of the neural network. Furthermore, it does not provide an optimization path for key parameters of the segmentation and detection process. We have developed techniques based on Bayesian inductive logic to extract specific parameter values for achieving optimal segmentation/feature extraction/identification of objects of interest, and our approach is applicable to the analysis of 1, 2 and 3 dimensional signals.

Our team has implemented a wide range of solutions to various government and military agencies in the fields of image manipulation, feature and object extraction, mapping applications, image compression, and data fusion.

Our strengths lie in our ability to recognize and extract objects and patterns in imagery using complex algorithms and neural networks; our innovative algorithm development for reconstructing 3D environments from various data sources, such as LADAR data; in creating intuitive mapping solutions for needs ranging from nautical to urban areas; and our highly effective Image Compression for various applications including transmission over low bandwidth channels.
>>>

Data Compression

Our research in the area of data compression focuses on achieving optimal compression for a wide variety of data types in a manner transparent to the end user.

We have developed unique algorithms for accomplishing both lossless and lossy compression that surpass the performance of existing data compression techniques. Our algorithms are incorporated into an overall compression methodology that serves as the foundation of our LPAC compression system.

The LPAC system is currently being utilized by the TEDServices initiative. The development of TEDServices has been undertaken by the U.S. Navy to “extend the Tactical Environmental Data Server (TEDS) into the era of Net Centric Warfare, Sea Power-21, FORCEnet, TaskForce Web, and the Navy Enterprise Portal (NEP).” An important goal of TEDServices is to deliver in an efficient manner a wide range of Meterological and Oceanographic (MetOc) data to end users. Data links for transmission have varying characteristics, ranging from 56kbps satellite links to 10Mbps LAN. LPAC is used to compress this data, using both lossless and lossy techniques as warranted, in a manner transparent to the end user. LPAC was evaluated against, and chosen over, a number of competing compression algorithms, including wavelet, JPEG2000, GRIB, and LZ77/78. Comparison metrics used in the comparison included compression ratio, quantitative loss (for lossy compression), CODEC speed, flexibility, and ease of use.
>>>