An Energy-Efficient Method for Human Activity Recognition with Segment-Level Change Detection and Deep Learning
| Autor: | Mooseop Kim, Chi Yoon Jeong |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
human activity recognition
Computer science Biosensing Techniques 02 engineering and technology lcsh:Chemical technology 01 natural sciences Biochemistry Convolutional neural network Article Analytical Chemistry Activity recognition 0202 electrical engineering electronic engineering information engineering Humans Context awareness Human Activities lcsh:TP1-1185 Electrical and Electronic Engineering Instrumentation segment-level change detection fully convolutional network business.industry Deep learning 010401 analytical chemistry deep learning Pattern recognition Energy consumption Atomic and Molecular Physics and Optics 0104 chemical sciences 020201 artificial intelligence & image processing Neural Networks Computer Artificial intelligence business energy-efficient method Algorithms Energy (signal processing) Change detection Efficient energy use |
| Zdroj: | Sensors Volume 19 Issue 17 Sensors, Vol 19, Iss 17, p 3688 (2019) Sensors (Basel, Switzerland) |
| ISSN: | 1424-8220 |
| DOI: | 10.3390/s19173688 |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
| Nepřihlášeným uživatelům se plný text nezobrazuje | K zobrazení výsledku je třeba se přihlásit. |