Author:
Isaeva Oksana L.,Boronenko Marina P.
Abstract
The prospects for using artificial intelligence in video analytics are becoming more and more undeniable. Researchers are actively working on methods for making decisions based on the results of automated analysis of the sequence of images received from video cameras. The purpose of our study was to develop a method for decision-making based on modeling a person's pupillary response to an information stimulus. A ZWO ASI120MC digital video camera was used for research. The studies carried out made it possible to reliably establish that the state of the optical system of the eye when viewing the calibration slide comes to a physical state common to all people. This made it possible to systematize the pupillograms, referring them to one of two categories (stress state is / is not). Having determined an individual threshold value during calibration, it becomes possible to quickly classify a person's emotional response to the received information impulse. Based on the developed methodology for systematizing the pupillary reaction to an information stimulus by categories, a decision tree for intelligent video surveillance systems was built. In addition to the size of the pupils, the algorithm takes into account the parameters of gaze fixation on stimulus images, the stability of the illumination of the surface of the pupils.
Reference14 articles.
1. Кольцов, А. С. Использование алгоритмов видеоаналитики в задачах выделения и распознавания человеческого лица в цифровом видеосигнале / А. С. Кольцов, Т. Е. Урусова, П. Ю. Гусев. – Текст : непосредственный // Техника и безопасность объектов уголовно-исполнительной системы. – Воронеж, 2018. – С. 33–35.
2. Бороненко, М. П. Телевизионная измерительная система наносекундного разрешения / М. П. Бороненко, П. Ю. Гуляев. – Текст: непосредственный // Доклады Томского государственного университета систем управления и радиоэлектроники. – 2014. – № 1 (31). – С. 60–64.
3. Каменева, М. Е. Психологические профили потенциально опасных лиц / М. Е. Каменева. – Текст : непосредственный // Вестник Московского университета МВД России. – 2010. – № 11. – С. 12–18.
4. Elevated response of human amygdala to neutral stimuli in mild post traumatic stress disorder: neural correlates of generalized emotional response / M. Brunetti [et al.] // Neuroscience. – 2010. – Т. 168, №. 3. – P. 670–679.
5. Machine learning to differentiate between positive and negative emotions using pupil diameter / A. Babiker [et al.] // Frontiers in psychology. – 2015. – № 6. – P. 19–21.