摘要： 在基于加速度传感器的人体行为识别中,分类器复杂度较高,易产生过拟合现象。为此,通过递阶遗传算法(HGA)训练BP神经网络作为分类器,采用三级染色体递阶结构表示神经网络的结构和参数。设计新的适应度函数,采用选择、交叉和变异操作联合优化BP网络的精确度和复杂度。测试结果表明,在基于加速度信号的行为识别系统中,相比基本HGA和其他常用算法,利用改进的HGA训练BP网络分类器可以有效控制网络结构,在保证隐层神经元数目较少的情况下,尽可能降低输出误差,实现两者的动态平衡,且对测试样本的识别正确率可达94.63%。

关键词: 行为识别, 加速度传感器, 递阶遗传算法, BP神经网络, 交叉, 变异

Abstract: The human activity recognition system based on accelerometer,referring to solve the problems such as high complexity and over fitting phenomenon,a Back Propagation(BP) neural network classifier trained via the Hierarchical Genetic Algorithm(HGA) is utilized.A three-layer chromosome hierarchical structure is used to optimize the structure and parameters of BP neural network simultaneously.A new fitness function is proposed,meanwhile,improved selection,crossover and mutation operator is beneficial to joint optimizing the complexity and accuracy of network.Results of tests show that it is better than the traditional HGA and other widely used algorithms in human body activity recognition system based on accelerometer.The BP neural network classifier based on HGA can effectively control the network structure and parameters.The average accuracy rate of test data is 94.63%.

中图分类号: 

• TP393