Activity and Emotion Recognition for Elderly Health Monitoring

Abstract

The problems facing elderly who are living independently, are considered to be one of the most important motivations of the activity recognition research. The advances in sensing technologies allow collecting several types of data and communicate it wirelessly. However, most existing activity recognition systems requires pre-calculated pattern recognition models. This paper lays the theoretical foundations of a real-time methodology for activity and emotional recognition based on body and environment sensors simultaneously, then tackles one aspect of the method which is the path estimation using chest-mounted IMU sensor, for which a zero velocity update criteria is proposed. Finally path estimation results for sitting down, laying down, falling down and standing up are discussed.

Emotion Recognition Techniques for Geriatric Users: A Snapshot

“…Shu et al [11] stated that physical signals such as facial expressions or speech are not heavily reliable as it is fairly easy for people to control outward physical expressions. This However, the features being used in most of current emotion recognition studies are still focusing on facial thermal imaging [18], [31]. They adopt temperature statistical features extracted from facial regions of interest or those commonly used in the visible spectrum.…”

“…A majority of emotion recognition studies focus of facial emotion recognition with only a few focusing on the elderly [14], [15], [18], [33], particularly with Alzheimer's for the elderly generation. They raised attention to the fact that while smart homes are intended to aid the elderly in with their daily needs, the technologies involved are unfamiliar to the elderly generation, due to which they are often uncomfortable with the implementations.…”

“…Benaissa et al [18] state that heart beat rate and breathing are used in various studies for emotion recognition as they are strong indicators of emotions. While other emotion recognition approaches are not very efficient in elderly monitoring, this method is more applicable in this case.…”

Novel tent pooling based human activity recognition approach

“…The ability to detect these movements is useful for a wide variety of applications. It is the most common method of detecting movements by attaching various motion sensors to the human body and the clothing they wear [11,22,58]. To use these methods, users must have fitted a computational radio frequency identification sensor [21].…”

Different Approaches for Human Activity Recognition: A Survey

“…Ronao & Cho [42] proposed a deep neural network based solution using embedded sensors in smartphone (accelerometer and gyroscope). Many other solutions have been presented for posture recognition using different wearable sensors [43], [44], [45], [46]. A major problem with wearable sensors is that it is not always feasible to wear these tags, while performing an activity.…”