巴西人群基本情绪面部表情识别面部数据库的开发与验证。
Development and validation of a face database for the recognition of facial expressions of basic emotions in the Brazilian population.
作者信息
de Morais Daiene, Maximiano-Barreto Madson Alan, Luchesi Bruna Moretti, Muniz Monalisa, Chagas Marcos Hortes Nisihara
机构信息
Department of Psychology, Federal University of São Carlos, São Carlos, SP, Brazil.
Study and Research Group in Mental Health, Cognition and Aging, University of São Paulo, Ribeirão Preto, Brazil.
出版信息
Psicol Reflex Crit. 2024 Dec 2;37(1):47. doi: 10.1186/s41155-024-00325-y.
BACKGROUND
The recognition of facial expressions of emotions is an essential skill for social functioning, as it enables recognizing the possible intentions of others.
MAIN BODY
Cultural context is an important aspect to consider in this skill, as it tends to modulate the recognition of facial expressions.
OBJECTIVE
To develop and validate a set of facial expressions of basic emotions of Brazilian individuals considering the population distribution of the country according to age group, sex, and race.
METHODS
A procedure with three phases was created to generate basic emotions and photograph facial expressions. A total of 459 Brazilians were then asked to identify the emotions of these facial expressions. Content validity was investigated based on the analysis of specialists and the recognition of emotions by Brazilian individuals.
RESULTS
The final database consists of 56 high-quality color images. A good level of agreement was found for the expressions perceived by the assessors (81.6%).
CONCLUSION
The percentage of correct recognition of each emotion and the characteristics of the models are presented so that future studies can choose the most adequate images to meet their specific needs.
背景
识别情绪的面部表情是社交功能的一项基本技能，因为它能够识别他人可能的意图。
主体内容
文化背景是这项技能中需要考虑的一个重要方面，因为它往往会调节对面部表情的识别。
目的
根据巴西按年龄组、性别和种族划分的人口分布情况，开发并验证一套巴西人的基本情绪面部表情。
方法
创建了一个分三个阶段的程序来生成基本情绪并拍摄面部表情照片。然后，共有459名巴西人被要求识别这些面部表情所表达的情绪。基于专家分析和巴西人对情绪的识别来调查内容效度。
结果
最终数据库由56张高质量彩色图像组成。评估者对所感知表情的一致性水平良好（81.6%）。
结论
给出了每种情绪的正确识别百分比和模型特征，以便未来的研究能够选择最合适的图像来满足其特定需求。
Zotero 插件