Coelho Fabrício, Pinto Milena F, Melo Aurélio G, Ramos Gabryel S, Marcato André L M
Federal University of Juiz de Fora (UFJF), Juiz de Fora, Brazil.
Federal Center for Technological Education of Rio de Janeiro (CEFET-RJ), Rio de Janeiro, Brazil.
Comput Methods Biomech Biomed Engin. 2023 Sep;26(9):1008-1017. doi: 10.1080/10255842.2022.2102422. Epub 2022 Jul 21.
The classification of sEMG signals is fundamental in applications that use mechanical prostheses, making it necessary to work with generalist databases that improve the accuracy of those classifications. Therefore, synthetic signal generation can be beneficial in enriching a database to make it more generalist. This work proposes using a variant of generative adversarial networks to produce synthetic biosignals of sEMG. A convolutional neural network (CNN) was used to classify the movements. The results showed good performance with an increase of 4.07% in a set of movement classification accuracy when 200 synthetic samples were included for each movement. We compared our results to other methodologies, such as Magnitude Warping and Scaling. Both methodologies did not have the same performance in the classification.
在使用机械假肢的应用中，表面肌电信号（sEMG）的分类至关重要，因此有必要使用能提高分类准确性的通用数据库。所以，合成信号生成有助于丰富数据库，使其更具通用性。这项工作提出使用生成对抗网络的一种变体来生成sEMG的合成生物信号。使用卷积神经网络（CNN）对动作进行分类。结果表明，当为每个动作包含200个合成样本时，一组动作分类准确率提高了4.07%，性能良好。我们将我们的结果与其他方法进行了比较，如幅度扭曲和缩放。这两种方法在分类中的表现并不相同。
