PROJECT: VISINT

Unique combination of biometric and bio-signal processing techniques, using neural networks (NN) for real-time analysis of human emotion and behavior.

PE1. Platform for automated assessment and interpretation of covert behavior for critical infrastructure personnel in a controlled area without the influence of external factors.

Platform for automated assessment and interpretation of covert behavior for critical infrastructure personnel in a controlled area without the influence of external factors (PE1). This platform is an analysis system that provides a continuous assessment of the ability of critical infrastructure (CI) personnel to perform specific tasks. Within this product there is a single person to track, placed in a controlled indoor environment.

Functionality

• Visual attention monitoring;
• Monitor the quality of the working environment (air, humidity and temperature conditions, noise level);
• Automatic identification of persons being assessed;
• Extract cues about the subject's behavior and affective experiences with deep learning solutions for: Face Detection (localization), Localization of important points on the face (e.g. nose tip, corners of the eyes, mouth, eyebrows, eyebrow corners, facial contour) and Facial Expression Recognition;
• Monitoring of relevant physiological signals (pulse, temperature, perspiration, respiration, blood oxygen level, ECG, EEG), provided reasonable intrusiveness is respected (sensors mounted on wristband, harness, inside of helmet, vest);
• Monitor movement parameters (acceleration, speed, shock, pedometer, GPS position, indoor relative position);
• Record intervention time;
• Real-time data transmission to a central monitoring station via radio interface;
• Real-time analysis of sensor data;
• The system will ensure the continuous recording of data and also flag deviations.

PE2. Platform for automated assessment and interpretation of covert behavior for critical infrastructure personnel in the open area with the influence of external factors

Platform for automated assessment and interpretation of dissimulated behavior for critical infrastructure personnel in the open area with the influence of external factors. This design variant is an analysis system that provides continuous assessment of the ability of CI personnel to perform specific tasks. This design variant considers an area covered by several surveillance cameras, where several persons (minimum 4 persons) are present.

Functionality

• The technical solution is implemented for a minimum of one group of four people;
• Continuous assessment of physiologic data with alarming for deviations;
• Training a (deep learning) system that will report behavioral cues based on the movement of different body parts (hand movements, head movements, displacements, etc.);
• Continuous evaluation of environmental and motion data, with deviation alarming, outdoor/indoor localization on a situation map;
• Analyze the recorded data, identify an individual response pattern and an overall acceptance norm.