%0 Journal Article %T 基于深度森林的下肢股直肌疲劳检测算法
Fatigue Detection Algorithm of Lower Limb Rectus Femoris Muscle Based on Deep Forest %A 卢小龙 %A 郭旭东 %A 郝又国 %A 高体杰 %A 王鹏 %A 周绣华 %J Modeling and Simulation %P 297-305 %@ 2324-870X %D 2022 %I Hans Publishing %R 10.12677/MOS.2022.112027 %X 肌肉疲劳评估是当今社会探索与研究的热点之一,与人们日常生活健康密切相关。目前常用的肌疲劳分类方法有支持向量机(Support Vector Machine, SVM)、随机森林(Random Forest, RF)和逻辑回归等浅层机器学习方法。为了进一步提高运动过程中肌肉疲劳识别的精度,因此,本文通过设计硬件采集装置进行下肢股直肌部位肌肉疲劳实验采集数据,构建深度森林(Deep Forest, DF)模型将预处理后的表面肌电(sEMG)数据分成三类:正常(normal)、疲劳(fatigue)和极度疲劳(extreme fatigue),并与传统方法SVM和RF模型对比。实验结果表明:深度森林模型的疲劳识别效果最好,整体准确率为92%,所提出的方法在各评价指标上均优于传统方法,基于深度森林的下肢股直肌疲劳检测算法,可以作为肌肉疲劳识别的方法,为人类健康提供参考与帮助。
Muscle fatigue assessment is one of the most popular research filed recently, and it is closely related to people’s daily health. At present, the commonly used classification methods of muscle fatigue include Support Vector Machine (SVM), Random Forest (RF), logistic regression and other shallow machine learning methods. In order to further improve the accuracy of muscle fatigue recognition, this paper designs a hardware acquisition device to collect data from the muscle fatigue experiment of the rectus femoris of the lower limbs, and constructs a deep forest model to divide the preprocessed surface electromyography (sEMG) data into three categories: normal, fatigue and extreme fatigue, compared with the traditional methods of SVM and RF models. The experimental results show that the deep forest model has the best fatigue recognition effect, with an overall accuracy rate of 92%. The proposed method is superior to traditional methods in all evaluation indicators. The deep forest-based fatigue detection algorithm for the rectus femoris of the lower limbs can be used as a method of muscle fatigue recognition to provide reference and help for human health. %K 肌疲劳,股直肌,sEMG,深度森林
Muscle Fatigue %K Rectus Femoris %K Surface Electromyography (sEMG) %K Deep Forest %U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=49158