Activity Recognition: Approaches, Practices and Trends

Abstract

Activity recognition has attracted increasing attention as a number of related research areas such as pervasive computing, intelligent environments and robotics converge on this critical issue. It is also driven by growing real-world application needs in such areas as ambient assisted living and security surveillance. This chapter aims to provide an overview on existing approaches, current practices and future trends on activity recognition. It is intended to provide the necessary material to inform relevant research communities of the latest developments in this field in addition to providing a reference for researchers and system developers who are working towards the design and development of activity-based context aware applications. The chapter first reviews the existing approaches and algorithms that have been used for activity recognition in a number of related areas. It then describes the practice and lifecycle of the ontology-based approach to activity recognition that has recently been under vigorous investigation. Finally the chapter presents emerging research on activity recognition by outlining various issues and directions the field will take. IntroductionWith the advance and prevalence of low-cost low-power sensors, computing devices and wireless communication networks, pervasive computing has evolved from a vision [1] to an achievable and deployable computing paradigm. As a result, research is now being conducted in all areas related to pervasive computing, ranging from low-level data collection, to intermediate-level information processing, to high-level applications and service delivery. It is becoming increasingly evident that intelligent environments which can support both living and work places through flexible multimodal interactions, proactive service provision, and context aware personalized activity assistance will be commonplace in the very near future. For example, Smart Homes (SH) -augmented real or simulated home settings equipped with sensors, actuators and information processing systems, have been extensively studied. Work in this area has produced a number of lab-based or real-world SH prototypes [2]. Within a SH the Activities of Daily Living (ADL) of its inhabitants, usually 1

Classification and suitability of sensing technologies for activity recognition

“…It comprises many di↵erent tasks, namely activity modelling, behaviour and environment monitoring, data processing and pattern recognition" [24]. There are many approaches for delivering AR which have been classified by various taxonomies -as illustrated by Fig.…”

“…2. One classification is based on the data type the AR system processes and thus there are two main classes: vision-based AR and sensor-based AR [99,24]. Vision-based AR uses visual sensing devices such as camera-based surveillance systems, or most recently, RGB depth cameras designed for motion-sensing applications and games such as Asus Xtion PRO LIVE 1 or Microsoft Kinect [138].…”

“…Sensor-based AR exploits sensor network technologies to measure actor's motions and their interaction with the environment. Chen and Khalil [24] further divide the sensor-based AR approaches into: wearable sensor-based and object-based (also called dense sensing). The former is achieved by attaching sensor(s) to an actor under observation -usually inertial measurement units, vital sign monitoring devices and Radio-Frequency Identification (RFID) tags or readers.…”

“…visual-based and sensor-based AR systems. Another classification is based on the type of recognition algorithms [24]: machine learning techniques and logical modelling & reasoning. Both taxonomies are valid and do not contradict each other, as one is based on the type of input and the other on the type of applied algorithms.…”

Rule-Based Real-Time ADL Recognition in a Smart Home Environment

“…Chen and Khalil [3] propose a broad categorisation into data-driven approaches, exploiting machine learning techniques, and knowledge-driven approaches, leveraging logical modelling and reasoning. Both directions have their strengths and weaknesses.…”