The presented study concerns development of a facial detection algorithm operating robustly in the thermal infrared spectrum. The paper presents a brief review of existing face detection algorithms, describes the experiment methodology and selected algorithms. For the comparative study of facial detection three methods presenting three different approaches were chosen, namely the Viola-Jones, YOLOv2 and Faster-RCNN. All these algorithms were investigated along with various configurations and parameters and evaluated using three publicly available thermal face datasets. The comparison of the original results of various experiments for the selected algorithms is presented.
This paper presents a study on the influence of psychophysical stimuli on facial thermal emissions. Two independent groups of stimuli are proposed to investigate facial changes resulting from human stress and physical exhaustion. One pertains to physical effort while the other is linked to stress invoked by solving a simple written test. Subjects’ face reactions were measured through collecting and analyzing long-wavelength infrared images. A methodology for numerical processing of images is proposed. Results of numerical analysis with respect to different facial regions of interest are provided. An automatic deep learning based algorithm to classify specific thermal face patterns is proposed. The algorithm consists of detection of regions of interests as well as numerical analysis of thermal energy emissions of facial parts. The results of presented experiments allowed the authors to associate emission changes in specific facial regions with psychophysical stimulations of the person being examined. This work proves high usability of thermal imaging to capture changes of heat distribution of face as reactions for external stimuli.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.