الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 166 |  |  |
| | | from | 166 | | |
Abstract
Upper limb prostheses are most significant for person amputees because they combine between essential functions of the natural hand or arm and cosmetic appearance. Previously prosthetic limbs were manufactured to simulate hand or arm movements normally. But many problems have appeared, including high cost. Also, the monopoly of companies manufactured and leading in the field of a prosthetic limb, causing many amputees unable to afford the initial cost for the installation of a prosthesis limb. All these problems made it difficult to obtain an artificial limb prosthetic. This thesis aims to realize the dream of persons to obtain suitable upper limb prosthetic hand, to achieve High-performance and low-cost through using a MATLAB program, artificial intelligence, programming, and 3D printer techniques. So Electromyography (EMG) signal was analyzed, processed for feature extraction and reduce noise using Daubechies wavelet transform and use many artificial intelligence techniques to Classify different movements. The output of signals classification was then used to distinguish between four various actions and control the hand movements. And have been integrated modern electronics (EMG sensor, Arduino programming) into the hand printed using the 3D printer. The performance of the hand designed theoretically and practically has also been verified, as will be shown later in this thesis. The results show that in order to classify these movements or patterns correctly, the basic features must be extracted using the (db8) Wavelet transformer Because they have led to better results in reducing the noise level and the Ensemble method has given The highest accuracy is 84.6% in Theoretical part, while the K-NN is the best classifier in the practical part where achieved The highest accuracy is 85% compared to the other types used.