改进的自适应遗传算法在人体行为识别中的应用

为提高传统自适应遗传算法优化的BP神经网络对人体行为的识别率,提出了一种改进的自适应遗传算法优化的BP神经网络预测方法.该算法使用新的动态变化的交叉和变异分布指数计算公式来优化传统的二进制交叉和多项式变异操作,根据种群集中和分散的剧烈程度自适应地增大或减小交叉和变异的概率,极大地弥补了传统的交叉和变异操作所造成的破坏优...

Application of Adaptive Genetic Algorithm in Human Behavior Recognition

In order to improve the recognition rate of human behavior by BP neural network optimized by traditional adaptive genetic algorithm,an improved BP neural network prediction method optimized by adaptive genetic algorithm is proposed.The algorithm optimizes the traditional binary crossover and polynomial mutation operations by using new dynamic crossover and mutation distribution index formulas.The probability of crossover and mutation can be increased or reduced adaptively according to the intensity of population concentration and dispersion,which greatly makes up for the defect of destroying excellent individuals caused by traditional crossover and mutation operations.By building the BP neural network model optimized by genetic algorithm,3000 groups of acceleration data of different human behaviors are tested.The experimental results show that the accuracy of the algorithm is improved from 92.31%to 96.47%.