Adaptive Thermal Sensor Array Placement for Human Segmentation and Occupancy Estimation

Abstract

Thermal sensor array (TSA) offers privacy-preserving, low-cost, and non-invasive features, which makes it suitable for various indoor applications such as anomaly detection, health monitoring, home security, and monitoring energy efficiency applications. Previous approaches to human-centred applications using the TSA usually relied on the use of a fixed sensor location to make the human-sensor distance and the human presence shape fixed. However, placing this sensor in different locations and new indoor environments can pose a significant challenge. In this paper, a novel framework based on a deep convolutional encoder-decoder network is proposed to address this challenge in real-life deployment. The framework presents a semantic segmentation of the human presence and estimates the occupancy in indoor-environment. It is also capable to segment the human presence and counts the number of people from different sensor locations, indoor environments, and human to sensor distance. Furthermore, the impact of the distance on the human presence using the TSA is investigated. The framework is evaluated to estimate the occupancy in different sensor locations, the number of occupants, environments, and human distance with classification and regression machine learning approaches. This paper shows that the classification approach using the adaptive boosting algorithm is an accurate approach which has achieves an accuracy of 98.43% and 100% from vertical and overhead sensor locations respectively.

Cooperative Networked PIR Detection System for Indoor Human Localization

“…Savazzi [ 10 ] proposed a Bayesian tool for tracking multiple bodies through real-time analysis of thermal signatures extracted from an 8 × 8 thermopile sensor array (i.e., 64 sensors), which monitors a 2.5 m square area when mounted on a 3.0 m ceiling and tracks people with an accuracy of about 0.5 m. However, considering the system complexity and the number of sensors used, the system may be limited to certain deployment scenarios in IoT applications. Naser [ 11 ] presented a framework based on a deep convolutional encoder-decoder network to perform semantic segmentation of the human presence and estimate the occupancy in the indoor environment. However, classification and regression machine learning approaches are required to perform human segmentation and occupancy estimation.…”

Multiple Thermal Sensor Array Fusion Toward Enabling Privacy-Preserving Human Monitoring Applications

“…Similarly, there has been recent works on using the TSA on human activity recognition, and abnormal behaviour detection [32]- [36]. The approach followed to process the TSA output is similar to image-processing approaches [37] while the analytical techniques on individual time intervals, frames, were different for instance Support Vector Machine (SVM) [12], Adaptive Boosting [6], [10], K-Nearest Neighbour (KNN) [30], [38], decision trees [20], [39], and Kalman filtering [40], [41]. One of the notable technical challenges reported in most human-based applications, which use TSA is that external heat sources have a major negative impact on the system performance.…”

“…Unlike conventional radars, Millimeter-Wave (mmWave) uses a short wavelength, which enables its radar to achieve high resolution and small antenna size, but makes it vulnerable to noise in indoor applications [5]. Third, vision-based sensing, for example, cameras that perform very well in real-world scenarios, although it violates users' privacy [6], clearly in domestic environments, e.g. homes, hospitals, nursing homes etc.…”

On the Gap between Domestic Robotic Applications and Computational Intelligence

“…Sensor-wise: sensors that use bio-metric information, such as cameras, sometimes can be replaced with other sensors such as LiDAR and infrared sensors [159][160][161]. If the bio-metric details of users must be captured, such details should be processed locally at the edge and should not be sent online.…”