Bloorview Research Institute and the Institute of Biomaterials and Biomedical Engineering, University of Toronto, Toronto, ON M4G 1R8, Canada.
IEEE Trans Biomed Eng. 2010 Apr;57(4):979-87. doi: 10.1109/TBME.2009.2035926. Epub 2009 Nov 17.
In this paper, time, frequency, and time-frequency features derived from thermal infrared data are used to discriminate between self-reported affective states of an individual in response to visual stimuli drawn from the International Affective Pictures System. A total of six binary classification tasks were examined to distinguish baseline and affect states. Affect states were determined from subject-reported levels of arousal and valence. Mean adjusted accuracies of 70% to 80% were achieved for the baseline classifications tasks. Classification accuracies between high and low ratings of arousal and valence were between 50% and 60%, respectively. Our analysis showed that facial thermal infrared imaging data of baseline and other affective states may be separable. The results of this study suggest that classification of facial thermal infrared imaging data coupled with affect models can be used to provide information about an individual's affective state for potential use as a passive communication pathway.
本文使用源自热红外数据的时间、频率和时频特征来区分个体对来自国际情感图片系统的视觉刺激的自我报告情感状态。共检查了六个二进制分类任务，以区分基线和情感状态。情感状态是根据被试报告的唤醒和效价水平来确定的。基线分类任务的平均调整准确率达到了 70%到 80%。唤醒和效价的高评级和低评级之间的分类准确率分别在 50%到 60%之间。我们的分析表明，基线和其他情感状态的面部热红外成像数据可能是可分离的。本研究的结果表明，面部热红外成像数据与情感模型的分类可用于提供个体情感状态的信息，可作为潜在的被动交流途径。
