人体生理特性和运动特性是影响步态识别的重要因素。利用实验采集的下肢表面肌电信号,首先对肌电信号进行小波消噪及特征提取,然后构造支持向量机分类器进行分类与识别,并针对步态周期数据的非均匀性(非等时性)特性进行了详细讨论。结果表明,即使在匀速行走条件下,人体步态周期仍然存在一定的非均匀特性,且这一特点会影响步态识别的准确性。这对于进一步研究步态稳定性和步态识别率等具有一定的参考价值。
The characteristics of the human physiology and motion are the important factors affecting the gait recognition. By means of the experimental data from the lower limb motion, firstly the surface electromyography (sEMG) was de-noised by the wavelet method and the feature samples were extracted, subsequently the classification and recognition were implemented by constructing the support vector machine (SVM)classifier, and the non-uniform(anisochronism) characteristics of the gait cycle were discussed in detail. The results show that even in a uniform walking condition, there still exit some non-uniform characteristics in the human gait cycle, which can affect the accuracy of the gait recognition. The work has a valuable reference to further study the gait stability and recognition rate.