Please use this identifier to cite or link to this item: http://dspace.ensta.edu.dz/jspui/handle/123456789/711
Full metadata record
DC FieldValueLanguage
dc.contributor.authorZITOUNI, Nour Elislam-
dc.contributor.authorSERAICHE, Oubada-
dc.contributor.authorREBAI, Karima (Directeur de thèse)-
dc.date.accessioned2026-11-19T09:02:10Z-
dc.date.available2026-11-19T09:02:10Z-
dc.date.issued2023-
dc.identifier.urihttp://dspace.ensta.edu.dz/jspui/handle/123456789/711-
dc.descriptionMémoire de fin d’étude de Master : Automatique et Informatique Industrielle: Alger, Ecole Nationale Supérieure des Technologie Avancées : 2023.en_US
dc.description.abstractLimb amputation reduces the patient's ability to perform daily living tasks. It is possible to handle this loss of autonomy by restoring certain functionalities of the amputated limb using a myoelectric prosthesis. This device performs various movements according to the muscular electrical activity of its user. It has been shown that a large majority of amputees can voluntarily generate muscle activation patterns that differ according to the movement they wish to perform. In addition, for actuation and reproducible movements of the upper limb, these activation patterns are also reproducible. They are thus a reliable and intuitive way to generate movements with a prosthesis. This is the best approach to providing the natural control wanted by these amputees. The development of a new method for the control of a myoelectric prosthesis involves the combination of several disciplines, such as human physiology, signal processing, and machine learning. Deep learning (DL) based solution has been used successfully in the development of intelligent systems in recent years and can effectively improve the quality of myoelectric prosthesis. It can be used for gestures recognition from surface Electromyogram (sEMG). The objective of this study is to compare different deep learning architectures for gesture recognition using surface electromyography (sEMG) signals in myoelectric prostheses. The aim is to identify the most effective architecture or combination of architectures to improve control and functionality for individuals with limb amputation.en_US
dc.language.isoenen_US
dc.publisherENSTAen_US
dc.subjectSurface EMGen_US
dc.subjectConvolution neural network Transfer learningen_US
dc.subjectrecurrent neural networken_US
dc.subjectprosthesis handen_US
dc.subjectfeatures extractionen_US
dc.subjecthand gesture recognitionen_US
dc.subjectSensor fusionen_US
dc.titleGesture Recognition Using Surface Electromyography and Deep Learning for Prostheses Hand: State-of-the-Arten_US
dc.typeThesisen_US
Appears in Collections:MFE- Automatique et Informatique Industrielle

Files in This Item:
File Description SizeFormat 
MFE 2023.AII.SERAICHE.Oubada_ZITOUNI.Nour.Elislam.pdf.pdfMémoire de Master525.22 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.