ROS System Facial Emotion Detection Using Machine Learning for a Low-Cost Robot Based on Raspberry Pi

5Citations
Citations of this article
24Readers
Mendeley users who have this article in their library.

Abstract

Facial emotion recognition (FER) is a field of research with multiple solutions in the state-of-the-art, focused on fields such as security, marketing or robotics. In the literature, several articles can be found in which algorithms are presented from different perspectives for detecting emotions. More specifically, in those emotion detection systems in the literature whose computational cores are low-cost, the results presented are usually in simulation or with quite limited real tests. This article presents a facial emotion detection system—detecting emotions such as anger, happiness, sadness or surprise—that was implemented under the Robot Operating System (ROS), Noetic version, and is based on the latest machine learning (ML) techniques proposed in the state-of-the-art. To make these techniques more efficient, and that they can be executed in real time on a low-cost board, extensive experiments were conducted in a real-world environment using a low-cost general purpose board, the Raspberry Pi 4 Model B. The final achieved FER system proposed in this article is capable of plausibly running in real time, operating at more than 13 fps, without using any external accelerator hardware, as other works (widely introduced in this article) do need in order to achieve the same purpose.

Cite

CITATION STYLE

APA

Martínez, J., & Vega, J. (2023). ROS System Facial Emotion Detection Using Machine Learning for a Low-Cost Robot Based on Raspberry Pi. Electronics (Switzerland), 12(1). https://doi.org/10.3390/electronics12010090

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free