Evaluation of convolutional neural networks for the classification of falls from heterogeneous thermal vision sensors

Abstract

The automatic detection of falls within environments where sensors are deployed has attracted considerable research interest due to the prevalence and impact of falling people, especially the elderly. In this work, we analyze the capabilities of non-invasive thermal vision sensors to detect falls using several architectures of convolutional neural networks. First, we integrate two thermal vision sensors with different capabilities: (1) low resolution with a wide viewing angle and (2) high resolution with a central viewing angle. Second, we include fuzzy representation of thermal information. Third, we enable the generation of a large data set from a set of few images using ad hoc data augmentation, which increases the original data set size, generating new synthetic images. Fourth, we define three types of convolutional neural networks which are adapted for each thermal vision sensor in order to evaluate the impact of the architecture on fall detection performance. The results show encouraging performance in single-occupancy contexts. In multiple occupancy, the low-resolution thermal vision sensor with a wide viewing angle obtains better performance and reduction of learning time, in comparison with the high-resolution thermal vision sensors with a central viewing angle.

Multi-Occupancy Fall Detection Using Non-Invasive Thermal Vision Sensor

“…Between the range of thermal vision sensors [25], the high resolution [26] and low resolution devices are used in smart environments [27]. In the case of fall detection, a comparative of thermal sensor devices [28] has shown that non-invasive and low resolution thermal sensors have better performance and reduction of learning time.…”

“…where ∆x = (∆x1,...,∆xd) ' , ∆xi, and i denote the stochastic perturbation, the stochastic perturbation on the i th input feature, and the number of input features, respectively. According to [28], for a given Q value, the upper bound of the LGEM (R * SM (Q)) is estimated by using the Hoeffiding's inequality with a probability of 1−η. The definition of R * SM (Q) is as follows:…”

Internet of things for health and well-being applications

“…Lo´pez-Medina et al 4 produced a solution which aims to detect falls in at-risk populations. This involved the use of privacy-preserving, low-resolution, thermal vision sensors intended to be mounted to the ceiling of a domiciliary area.…”

An Infrared Array Sensor-Based Approach for Activity Detection, Combining Low-Cost Technology with Advanced Deep Learning Techniques

“…Miguel et al [ 21 ] proposed a fall detection system using two IR array sensors: one having a High Resolution (HR) and the other having a Low Resolution (LR). The collected thermal data are in a fuzzy representation; the activities are classified using CNNs and the achieved accuracy is equal to 94.3%.…”