一种轻量化的CNN活动识别方法研究

Lightweight CNN Activity Recognition Algorithm

目前大多数人类活动识别研究方法需要人工提取传感器数据特征,将特征值输入到分类器中完成识别,在提取特征值过程中可能会丢失大量的信息,导致活动识别率受到影响.本文提出一种轻量化的卷积神经网络(Convolutional Neural Networks,CNN)活动识别方法,首先将加速度数据转化为图像,然后构建CNN模型并研究训练模型压缩方法,最后将图像输入到CNN模型中识别人体的具体活动.实验结果表明,对6种人类活动的平均识别率达到97. 6%,高于目前提出的大部分活动识别方法.

At present,most human activity recognition research methods need to extract sensor data features,input feature values into the classifier to complete the recognition,and a large amount of information may be lost in the process of extracting feature values,resulting in the activity recognition rate being affected. This paper proposes a lightweight Convolutional Neural Networks( CNN) activity recognition method,which first converts the acceleration data into images,then constructs the CNN model and studies the training model compression method,and finally inputs the image into the CNN model. Identify the specific activity of the human body. The experimental results show that the average recognition rate of the six human activities is 97. 6%,which is higher than most of the current activity recognition methods.