Periorbital thermal signal extraction and applications

Area of Science:

• Physiological measurement
• Biomedical engineering
• Computational psychology

Background:

• The periorbital signal, reflecting blood supply to ocular muscles, is linked to autonomic responses and stress.
• Previous methods for extracting this signal used heuristic segmentation, limiting accuracy and consistency.
• Accurate periorbital signal extraction is crucial for stress quantification and applications like deception detection.

Purpose of the Study:

• To develop and validate a novel, accurate, and consistent method for localizing and extracting the periorbital thermal signal.
• To replace existing heuristic segmentation approaches in stress quantification.
• To advance applications in computational psychology, particularly in deception detection.

Main Methods:

• Localization of the thermal footprint of facial and ophthalmic arterial-venous complexes in the periorbital area.
• Extraction of the mean thermal signal over time (periorbital signal) from the localized footprint.
• Testing the method on thermal videos of 39 subjects undergoing stressful interrogation for a mock crime.

Main Results:

• The proposed method enables accurate and consistent extraction of the periorbital signal.
• The new approach significantly improved deception classification success rate to 82%.
• This represents a 20% increase in classification accuracy compared to previous methods.

Conclusions:

• The novel thermal signal localization and extraction method offers a significant improvement over heuristic approaches for stress quantification.
• This technology has direct applications in enhancing the accuracy of deception detection systems.
• The improved accuracy in deception classification highlights the potential of this method in forensic and psychological research.