Photoheads is a concept created by Dr. Terry Boult to create datasets by re-imaging synthetic data, such as images. Using synthetic data allows for datasets to be captured in situations where using human subjects would be difficult or not feasible.
FAMME: Facial Analysis in Moving Maritime Environments is a Small Business Innovative Research (SBIR) contract in conjunction with Securics, Inc. developing a full end-to-end face recognition pipeline for unconstrained environments and, specifically, for maritime use. Methods developed include deblurring techniques for motion and atmospheric aberrations, as well as integration with robust recognition cores.
The VAST Lab is collaborating with the University of Maryland, Columbia University, University of California (San Diego), University of Southampton, and the University of Queensland on a Multi-University Research Initiative (MURI) on multiple methods of determining identity in (primarily) ship-to-ship situations. Our particular portion of the MURI entails performing face recognition at range using embedded systems and is closely aligned with the efforts in the FAMME project, though not overlapping.
Current biometric methods jeopardize privacy by imposing a danger of biometric data becoming permanently compromised. This is because, unlike a password or Social Security Number, it cannot be changed. Dr. Boult has developed an approach that allows biometric data to be converted to a secure but revocable form that still allows the computation of robust distance needed for effective biometric data. The result is a technique that preserves privacy but can enhance security.
This is an NIMH funded project with subcontracts between the University of Colorado at Colorado Springs Vision and Security Technology Laboratory, the Psychology Department, and Blue Sun Incorporated.
This project is looking at novel imaging systems and techniques to increase the effective dynamic range, i.e. the range of brightest to darkest measurements in an image. The project is developing both new imaging system using optics allowing per-pixel modulation of the incoming light, control systems to define the modulation, and software techniques to fuse multiple spatio-temporal samples.
The Catadioptric Canvas project is developing software for designing new optical systems that combine novel shaped mirrors with lenses to allow almost any user-defined mapping between the world and the imaging sensor.
There is always news on terrorism and terrorist activities around the world. These current circumstances have lead us to such a stage that it has become absolutely essential to devise methods of detecting anti-social activities at an early stage. This project will be catering to the needs of the Northern Command, to come up with a method of detecting Chemical and Biological warfare agents in maritime cargo.
A subcontract on a Department of Education Phase I SBIR to Active Artz.
A NISSC funded project joint with X. Zhou of UCCS CS Department.
DARPA HID program contract number N00014-00-1-0929.
This effort would develop a system that would address two areas of concern in tracking the location and coordinating timing of events for trauma patients. The effort will develop a prototype of a wearable software defined radio which can be attached to a trauma patient upon first contact by medical staff. NAVSYS is the project lead doing the system design, and UCCS is analyzing the wireless networking options.
ONR funded Phase 1 STTR project joint with Stream Processors Inc (SPI is a Stanford University Spin-off).
Subcontract on a ONR Phase II SBIR to Navsys Inc.
This research work is part of ongoing project by Dr Boult, The El Pomar Endowed Chair of Communication and Computation in the Computer Science Dept, UCCS. He was the Chief Technology Officer (CTO) of Guardian Solutions Inc when the project was started. The project, however, is done entirely using university funds.
Video Surveillance is becoming ubiquitous, except in those locations where privacy truly dominated. Unfortunately, surveillance may simply serve to displace crime.
Joint work with C. Wang and Gang Zhang UCCS ECE. Partially funded by NISSC last year. Recently filed preliminary patent, currently working on spin-off company.
SEE Port: Surveillance and Enemy Evaluation with PTZ and Omni-directional Reconnaissance Technology
Subcontract on a ONR Phase II SBIR to Remote Reality Inc.
Networked Enhanced Automated Target Recognition (NEATR), is an important problem in modern network centric operations. This Phase I effort focuses on fusion in an embedded system with reasonable Size Weight and Power (SWAP).
Stegi@Work leverages idle processing power within an organization to provide compute intensive steganography detection tasks. Subscribed users submit files to the system for analysis and a series of volunteer systems within multiple priority and security queues provide the CPU power to process the task.
Configuring the IQEye3 network camera for use in the DARPA grand challenge project.
The Verilog VPI is a C-Programming interface to the Verilog HDL (Hardware Description Language). It essentially allows you to write your own Verilog simulator system calls using C/C++.