A Two-stage Method for Solving Multi-resident Activity Recognition in Smart Environments

Abstract

Abstract:To recognize individual activities in multi-resident environments with pervasive sensors, some researchers have pointed out that finding data associations can contribute to activity recognition and previous methods either need or infer data association when recognizing new multi-resident activities based on new observations from sensors. However, it is often difficult to find out data associations, and available approaches to multi-resident activity recognition degrade when the data association is not given or induced with low accuracy. This paper exploits some simple knowledge of multi-resident activities through defining Combined label and the state set, and proposes a two-stage activity recognition method for multi-resident activity recognition. We define Combined label states at the model building phase with the help of data association, and learn Combined label states at the new activity recognition phase without the help of data association. Our two stages method is embodied in the new activity recognition phase, where we figure out multi-resident activities in the second stage after learning Combined label states at first stage. The experiments using the multi-resident CASAS data demonstrate that our method can increase the recognition accuracy by approximately 10%.

Modeling interaction in multi-resident activities

“…On the contrast, some investigations (Alemdar et al, 2013) (Chen and Tong, 2014) ) did consider interaction between residents. In particular, (Alemdar et al, 2013) (Chen and Tong, 2014) used conventional HMM and CRF showing the advantage of fitting such classical graphical models for single resident activity recognition.…”

“…In the inference step, the combined label state is inversely mapped onto the corresponding individual activity labels. (Chen and Tong, 2014) compared the HMM-based method against the one applied in on the same dataset, showing that their method was on average better by almost 10% and that HMM performed slightly better than CRF.…”

“…In particular, (Alemdar et al, 2013) (Chen and Tong, 2014) used conventional HMM and CRF showing the advantage of fitting such classical graphical models for single resident activity recognition. They used respectively a single HMM chain and a single CRF chain to model multi-resident activities by defining "combined labels".…”

“…− We propose a variant of the combined label approach based on HMM applied in (Chen and Tong, 2014) (Alemdar et al, 2013), we call it CL-HMM. In addition to the use of combined labels for the pair of activities labels (i.e., Resident 1 activity label and Resident 2 activity label), our approach also suggests the use of combined labels for the pair of observations (i.e., the observation of Resident 1 and the observation of Resident 2).…”

“…In , Coupled Hidden Markov Model (CHMM) was applied. In (Chen and Tong, 2014) both HMM and Conditional Random Field (CRF) were applied and compared. In (Alemdar et al, 2013) HMM was used to model with multi-resident activities.…”