Screen use can influence the circadian phase and cause eye strain. Smart eyeglasses with an integrated color light sensor can detect screen use. We present a screen use detection approach based on a light sensor embedded into the bridge of smart eyeglasses. By calculating the light intensity at the user's eyes for different screens and content types, we found only computer screens to have a significant impact on the circadian phase. Our screen use detection is based on ratios between color channels and used a linear support vector machine to detect screen use. We validated our detection approach in three studies. A test bench was built to detect screen use under different ambient light sources and intensities in a controlled environment. In a lab study, we evaluated recognition performance for different ambient light intensities. By using participant-independent models, we achieved an ROC AUC above 0.9 for ambient light intensities below 200 lx. In a study of typical ADLs, screen use was detected with an average ROC AUC of 0.83 assuming screen use for 30% of the time.

No abstract

Numerous approaches exist for disaggregating power consumption data, referred to as non-intrusive load monitoring (NILM). Whereas NILM is primarily used for energy monitoring, we intend to disaggregate a household’s power consumption to detect human activity in the residence. Therefore, this paper presents a novel approach for NILM, which uses pattern recognition on the raw power waveform of the smart meter measurements to recognize individual household appliance actions. The presented NILM approach is capable of (near) real-time appliance action detection in a streaming setting, using edge computing. It is unique in our approach that we quantify the disaggregating uncertainty using continuous pattern correlation instead of binary device activity states. Further, we outline using the disaggregated appliance activity data for human activity recognition (HAR). To evaluate our approach, we use a dataset collected from actual households. We show that the developed NILM approach works, and the disaggregation quality depends on the pattern selection and the appliance type. In summary, we demonstrate that it is possible to detect human activity within the residence using a motif-detection-based NILM approach applied to smart meter measurements.

Human activity event recognition (HAER) within a residence is a topic of significant interest in the field of ambient assisted living (AAL). Commonly, various sensors are installed within a residence to enable the monitoring of people. This work presents a new approach for HAER within a residence by (re-)using measurements from commercial smart water meters. Our approach is based on the assumption that changes in water flow within a residence, specifically the transition from no flow to flow above a certain threshold, indicate human activity. Using a separate, labeled evaluation data set from three households that was collected under controlled/laboratory-like conditions, we assess the performance of our HAER method. Our results showed that the approach has a high precision (0.86) and recall (1.00). Within this work, we further recorded a new open data set of water consumption data in 17 German households with a median sample rate of 0.083¯ Hz to demonstrate that water flow data are sufficient to detect activity events within a regular daily routine. Overall, this article demonstrates that smart water meter data can be effectively used for HAER within a residence.

No abstract