Selección de características usando el algoritmo LEM para la clasificación de señales EMG

In medical applications, the amputation of an arm or the lack of a limb of the body inspires the technological advances in the area of robotics for the creation of intelligent prosthesis replaces and recovers a percentage of the functionality of the absent limb of a person. One of the most important bases for the development of robotic limbs is the analysis and study of EMG signals (surface electromyographic signals). EMG signals rovide information on the dynamics of a muscle in its different states and provide amplitude and frequency values that describes the movement, contraction and rest of a muscle. For an EMG signal, there are representative characteristics like the RMS value, Histogram, standard deviation, among other functions that allow characterizing a given signal in the time domain and frequency. The objective is to compare the most commonly used approaches and characteristics of EMG signals to differentiate between different signals that represent gestures or movements of the hand.

• Autores: Londoño Lopera, Juan Camilo; González Alzate, Juan Pablo; Lage Cano, Esteban Camilo; Vallejo Velasquez, Mónica Ayde; Ramírez Patiño, Juan Fernando
• Fecha de publicación: 2020
• Institución: Universidad Santo Tomás
• Repositorio: Universidad Santo Tomás
• Idioma: spa
• OAI Identifier: oai:repository.usta.edu.co:11634/26845
• Acceso en línea: http://revistas.ustatunja.edu.co/index.php/ingeniomagno/article/view/1896; http://hdl.handle.net/11634/26845
• Palabra clave: Electromyography; classifier; gesture recognition; evolutionary algorithm; prosthesis; robotic hand; Electromiografía; clasificador; reconocimiento de gestos; algoritmo evolutivo; prótesis; mano robótica; Eletromiografia; classificador; reconhecimento de gestos; algoritmo evolutivo; prótese; mão robótica
• License: Derechos de autor 2020 Ingenio Magno