Irvine Sensors Corporation

Isc8 Inc.

Virgilio Villacorta

James W Justice

Ying Wen Hsu

W Eric Boyd

David E Ludwig

Medhat Azzazy

Fredrik Knutson

Mannchuoy Yam

Omar Asadi

Eric Weaver

The Hy-ALERT processing system operates simultaneously on the data streams from visible and thermal imaging sensors and from hyperspectral imaging sensors. The visible and thermal image processing uses cognitive-inspired processing techniques that determine the salient content of the imagers based on the system user definition of salient content. The hyperspectral image processing uses anomaly detection and template matching techniques to determine the salient materials content of the observed scenes. Cross-correlation processing of the salient content of the various image and hyperspectral data streams enables the sensor systems to adapt their operating characteristics to optimize effectiveness in near real-time.

Real time signal processor for analyzing, labeling and exploiting data in real time from hyperspectral sensor suites (Hy-ALERT)

An invention is disclosed for a real time video analytic processor that embodies algorithms and processing architectures that process a wide variety of sensor images in a fashion that emulates how the human visual path processes and interprets image content. spatial, temporal, and color content of images are analyzed and the salient features of the images determined. These salient features are then compared to the salient features of objects of user interest in order to detect, track, classify, and characterize the activities of the objects. Objects or activities of interest are annotated in the image streams and alerts of critical events are provided to users. Instantiation of the cognitive processing can be accomplished on multi-FPGA and multi-GPU processing hardwares.

Methods and devices for cognitive-based image data analytics in real time

The invention comprises a lip reading device having a capacitive array for enhanced portable speech recognition. The capacitive array of the invention produces a sequence of signal frames or signal data sets (i.e., digitized output) representative of the proximity and motion of a user's lips at a predetermined sample rate and resolution. The sequence of signal data sets is stored in a first electronic memory and are compared against a reference data set representative of a predetermined acoustic signal stored in a second electronic memory. The attributes of signal data set are compared against the reference data set for likely data matches based on predetermined criteria.

Portable lip reading sensor system

Growing community of inventors

Corona, CA, United States of America

Virgilio Villacorta