基于SVM_KNN的老人跌倒检测算法

Fall Detection Algorithm Based on SVM_KNN

跌倒是老年人伤害和死亡的主要诱因之一,我国每年约有4000万65岁以上的老人意外跌倒。本文基于智能手机的加速度、气压计等传感器提出一种人体跌倒检测算法。该算法首先采用支持向量机(SVM)对训练集进行训练,得到一个弱二分类器(包含最优超平面和支持向量集),然后计算待测样本到最优超平面的距离。若该距离大于设定的间隔,直接采用SVM分类;否则,利用支持向量集作为有标签的训练集进行K近邻分类(KNN)。考虑到特征值的多维性,本文引入标准化欧氏距离替代传统的欧氏距离。仿真与实验结果显示,与传统的支持向量机算法相比,该算法能有效提高跌倒检测的准确率,且不受智能手机放置位置的限制。

Falling is one of the main causes of casualties in the elderly, every year about accidentally. To improve the accuracy in human fall detection, a fall detection algoritlim based on accter in a smart phone is proposed, the algoritlim is an improved support vector machine （the training set to obtain a weak 2-classifier （including the optimal hyperplane andistancc from the sample to the optimal hyperplane. I the distance is greater than the givclassified with SVM. Otherwise, the K-nearest-neighbor classifier （ KNN） method will be used. In addition, in the KNNmethod,the distancc between the eigenvectors is calculated using the standard Euclidean distance. Swith the non-optimized support vector machine algorithm, this algorithm can effectively improve the fall detection accuracy andsmartphones can be placed casually.