Abstract
Human activity recognition (HAR), which is important in context awareness services, needs to occur continuously in daily life, owing to which an energy-efficient method is needed. However, because human activities have a longer cycle than HAR methods, which have analysis cycles of a few seconds, continuous classification of human activities using these methods is computationally and energy inefficient. Therefore, we propose segment-level change detection to identify activity change with very low computational complexity. Additionally, a fully convolutional network (FCN) with a high recognition rate is used to classify the activity only when activity change occurs. We compared the accuracy and energy consumption of the proposed method with that of a method based on a convolutional neural network (CNN) by using a public dataset on different embedded platforms. The experimental results showed that, although the recognition rate of the proposed FCN model is similar to that of the CNN model, the former requires only 10% of the network parameters of the CNN model. In addition, our experiments to measure the energy consumption on the embedded platforms showed that the proposed method uses as much as 6.5 times less energy than the CNN-based method when only HAR energy consumption is compared.
Highlights
The recognition of human activity plays a vital role in context awareness services such as daily lifelogging, surveillance, healthcare, and human-computer interaction
We proposed an energy-efficient method for the overall system of human activity recognition
Because human activities have a longer cycle than Human Activity Recognition (HAR) methods, which perform their analysis in cycles of a few seconds, classifying human activities continuously by using a short cycle is highly inefficient
Summary
The recognition of human activity plays a vital role in context awareness services such as daily lifelogging, surveillance, healthcare, and human-computer interaction. Previous research on HAR using smartphones mainly involved two approaches: methods based on handcrafted features and deep-learning-based methods. The disadvantage of DNNs is their high computational costs to achieve high accuracy This limitation would render them unsuitable for deployment on embedded devices, which would require the energy consumption of the deep-learning methods used for HAR to be considered. HAR methods based on handcrafted features mainly reduce the energy consumption by lowering or varying the sampling rate of the inertial sensors [12,13]. Several methods using shallow networks to reduce the energy consumption of activity recognition engines based on DNNs were proposed [14,15,16]. We proposed an energy-efficient method for human activity recognition with segment-level change detection and deep learning.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
