Camera-based facial expression recognition system to analyze customer satisfaction level in a restaurant

• Autor: Lara-Lévano, Edwin
• Formato: artículo
• Fecha de Publicación: 2019
• Institución: Universidad de Lima
• Repositorio: Revistas - Universidad de Lima
• Lenguaje: español
• OAI Identifier: oai:ojs.pkp.sfu.ca:article/4638
• Enlace del recurso: https://revistas.ulima.edu.pe/index.php/Interfases/article/view/4638
• Nivel de acceso: acceso abierto
• Materia: Customer satisfaction; facial expression recognition; histogram of oriented gradients; support vector machine; facial landmarks; Satisfacción del cliente; reconocimiento de expresiones faciales; histograma de gradientes orientados; máquina de soporte vectorial; puntos de referencia faciales

The main objective of this research is to develop a system that recognizes customers’ satisfaction or dissatisfaction in a restaurant based on their facial expressions when receiving the service provided by the establishment. The implementation of the system has a series of stages common to the development of a visual computing project, which will begin with data preprocessing to train the classifier to be used in this case: a support vector machine. This data preprocessing uses the histogram of oriented gradients for detecting a face inside an image, so that only the face outline is cut. In this way, facial landmarks of the image are extracted, the probability that each of the basic feelings appears in the facial expression of people is established, and, based on these probabilities, customers’ satisfaction or dissatisfaction is determined. The results show that the system correctly detected most of the images entered for the tests; however, there were some cases in which, despite the fact that customers were satisfied, they showed certain dissatisfaction expressions caused by external factors.