Activity identification using body-mounted sensors—a review of classification techniques

Abstract

With the advent of miniaturised sensing technology, which can be body-worn, it is now possible to collect and store data on different aspects of human movement under the conditions of free-living. This technology has the potential to be used in automated activity profiling systems which produce a continuous record of activity patterns over extended periods of time. Such activity profiling systems are dependent on classification algorithms which can effectively interpret body-worn sensor data and identify different activities. This article reviews the different techniques which have been used to classify normal activities and/or identify falls from body-worn sensor data. The review is structured according to the different analytical techniques and illustrates the variety of approaches which have previously been applied in this field. Although significant progress has been made in this important area, there is still significant scope for further work, particularly in the application of advanced classification techniques to problems involving many different activities.

Assessing sedentary behavior using wearable devices: An overview and future directions

“…Overviews of feature extractions and activity recognitions were given by Liu et al 36) and Preece et al 37) , respectively. The readers can refer to these articles for further information.…”

Coarse In-Building Localization with Smartphones

“…It promises to have great impact on a variety of domains like elder care [6], fitness monitoring [9,14], and intelligent contextaware applications [5]. Recently several researchers have used the accelerometers embedded in smartphones to detect activities such as walking, standing, running and sitting with the goal of developing context-aware applications [8,10].…”

Fuzzy Inference System for Physical Activity Recognition Based on General Features and PCA

“…Physical activity monitoring can be used for healthcare services for patients and elder people [1]. Practical applications include falling detection, energy expenditure monitoring and physical therapy management.…”

“…Various methods have been proposed for physical activity recognition, which can be found in two reviews of [1], [3]. Karantonis, Mathie, et al used one waist-worn accelerometer and a real-time threshold-based classification algorithm to recognize daily activities.…”

“…To determine the activity type in real-time way, the acceleration signal samples are segmented into window series and only newest 50 acceleration signal samples are used for classification for each window. 1 The number 50 means time window length of one second, which is enough to include one typical movement cycle such as one walking step.…”