改进的PSO-SVM在表面肌电信号模式识别中的研究

A Support Vector Machine Based on an Improved Particle Swarm Optimization Algorithm for SEMG Signal Pattern Recognition

为了提高表面肌电信号的遥操作机械手运动模式识别率,设优化支持向量机(IPSO-SVM)。该方法首先简化PSO的位置和速度公式,然后提出ESE状态估计策略判断算法的"早熟"收敛,最后对6类手臂运动模式(握拳、展拳、内旋、外旋、屈腕、伸腕)进行分类并与另外4个测试算法的分类结果进行比较。实验结果表明:IPSO-SVM算法的平均准确率为93.75%,而传统SVM算法的平均准确率为70.21%;算法的训练时间和泛化时间都有明显的提高;具有较强的鲁棒性和抗干扰能力。因此IPSO-SVM算法可以很好的解决表面肌电信号的动作模式分类问题,具有很好的应用价值。

In order to improve the motion pattern recognition rate of EMG signals,this paper proposes an improved PSO algorithm to optimize SVM（ IPSO-SVM）. Firstly,IPSO-SVM introduces a way to simplify the position and velocity formulas of PSO,then proposes ESE state estimation for premature convergence,and finally adopts 5 test algorithms to classify the six hand motion patterns recognition（ fist clenching,fist unfolding,internal and external rotation,wrist intorsion and wrist extorsion）. The results showed that the average accuracy rate of IPSO-SVM is 93.75%and the average accuracy of traditional SVM algorithm is 70.21%; the training and testing time were also obviously reduced. It also has strong robustness and noise immunity. Therefore,the IPSO-SVM algorithm can be used to solve the classification problem of the surface EMG signal,which has a good application value.