This awardee has received supplemental funding. This award detail page includes information about both the original award and supplemental awards.
Description of original award (Fiscal Year 2010, $409,226)
StereoVision Imaging, Inc. (SVI) proposes to develop and demonstrate a cost-effective, mobile wireless device capable of verifying a biometric at a distance of up to 1000 meters with a hand held binocular system under uncontrolled lighting and environmental conditions. SVI proposes using existing 2D interagency facial databases with SVI's data rich 3D facial recognition technology (FRT) to accomplish the objective.
StereoVision Imaging, Inc. will complete their prototype mobile facial recognition system that includes surveillance binoculars with on-demand facial recognition and system integration with legacy 2D database images of persons of interest. Face recognition from uncooperative individuals in ambient environments remains the most challenging surveillance scenario. This application leverages previous NIJ investments (awards 2009-IJ-CX-K001 and 2010-IJ-CX-K001) to enhance existing facial recognition surveillance systems using an innovative technological approach to raise the bar toward meeting one of the most demanding technological challenges for the criminal justice practitioner. The current prototype device captures images with manual settings for lighting and focus at specific ranges of 50, 75, and 100 meters. The added utility of auto-focus, auto-exposure, and shutter has been found to be critical to advancing image quality across the spectrum of variable range and ambient outdoor lighting conditions thus providing for quick set-up and use. SVI proposes to increase the device ease of use and performance by incorporating:
1. An auto-exposure algorithm
2. An auto-focus algorithm
3. Blur reduction techniques through global shutter versus current rolling shutter implementation
4. An upgraded commercial off-the-shelf (COTS) face matcher application specifically optimized for the StereoVision image collection and segmentation application.