Elderly Monitoring and Action Recognition System Using Stereo Depth Camera

Zin, Thi Thi; Htet, Ye; Akagi, Yuya; Tamura, H.; Kondo, Kazuhiro; Araki, S.

doi:10.1109/gcce50665.2020.9291785

Cited by 3 publications

(7 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experiment tested and evaluated this system using a dataset that was collected at an elder care center. As for feature vectors, appearance-based features [ 24 ] (depth motion appearance and depth motion history with HOG descriptor) are used in combination with the distance-based features [ 21 ] (3D human centroid height relative to the floor) and fused together with the automatic rounding method [ 28 ] for action recognition. The duration that forms a specific action is automatically annotated by the rounding method, and the action is recognized by an SVM classifier according to the feature vector.…”

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…Many algorithms have been proposed to recognize actions using representations built from 3D silhouettes [ 6 , 21 ], as well as using depth motion maps [ 22 ] obtained from 3D silhouettes. Among these, a threshold-based action recognition algorithm using distance features from 3D silhouettes [ 21 ] was recently proposed. In this system, the 3D human centroid height relative to the floor is used as a feature for action recognition, and the percentage of body movement from frame differencing is used for sleep monitoring.…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, this method is not useful for any indoor environment at elder care centers, as the depth images provided by the proposed depth camera have large depth noise variations, and thus, the pre-defined threshold values will differ from one indoor environment to another. To address this issue, this previous action recognition system [ 21 ] is extended in our proposed system by extracting the spatial-temporal features from the depth maps which are fused together with distance features to improve accuracy and reliability.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Real-Time Action Recognition System for Elderly People Using Stereo Depth Camera

Zin

Htet

Akagi

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

Smart technologies are necessary for ambient assisted living (AAL) to help family members, caregivers, and health-care professionals in providing care for elderly people independently. Among these technologies, the current work is proposed as a computer vision-based solution that can monitor the elderly by recognizing actions using a stereo depth camera. In this work, we introduce a system that fuses together feature extraction methods from previous works in a novel combination of action recognition. Using depth frame sequences provided by the depth camera, the system localizes people by extracting different regions of interest (ROI) from UV-disparity maps. As for feature vectors, the spatial-temporal features of two action representation maps (depth motion appearance (DMA) and depth motion history (DMH) with a histogram of oriented gradients (HOG) descriptor) are used in combination with the distance-based features, and fused together with the automatic rounding method for action recognition of continuous long frame sequences. The experimental results are tested using random frame sequences from a dataset that was collected at an elder care center, demonstrating that the proposed system can detect various actions in real-time with reasonable recognition rates, regardless of the length of the image sequences.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Real-Time Action Recognition System for Elderly People Using Stereo Depth Camera

Zin

Htet

Akagi

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…They utilized trained ML classifiers to enable smart wards and health monitoring systems. Furthermore, Zin et al [5] proposed a stereotype depth camera for monitoring the sleep of patients. They utilized this sleep data to examine their health.…”

Section: Related Work a Health Monitoring Systemsmentioning

confidence: 99%

i-Sheet: A Low-Cost Bedsheet Sensor for Remote Diagnosis of Isolated Individuals

2023

View full text Add to dashboard Cite

In this article, we propose a smart bedsheeti-Sheet-for remotely monitoring the health of COVID-19 patients. Typically, real-time health monitoring is very crucial for COVID-19 patients to prevent their health from deteriorating. Conventional healthcare monitoring systems are manual and require patient input to start monitoring health. However, it is difficult for the patients to give input in critical conditions as well as at night. For instance, if the oxygen saturation level decreases during sleep, then it is difficult to monitor. Furthermore, there is a need for a system that monitors post-COVID effects as various vitals get affected, and there are chances of their failure even after the recovery. i-Sheet exploits these features and provides the health monitoring of COVID-19 patients based on their pressure on the bedsheet. It works in three phases: 1) sensing the pressure exerted by the patient on the bedsheet; 2) categorizing the data into groups (comfortable and uncomfortable) based on the fluctuations in the data; and 3) alerting the caregiver about the condition of the patient. Experimental results demonstrate the effectiveness of i-Sheet in monitoring the health of the patient. i-Sheet effectively categorizes the condition of the patient with an accuracy of 99.3% and utilizes 17.5 W of the power. Furthermore, the delay involved in monitoring the health of patients using i-Sheet is 2 s which is very diminutive and is acceptable.

show abstract

HMM-Based Action Recognition System for Elderly Healthcare by Colorizing Depth Map

Htet

Zin

Tin

et al. 2022

IJERPH

Self Cite

View full text Add to dashboard Cite

Addressing the problems facing the elderly, whether living independently or in managed care facilities, is considered one of the most important applications for action recognition research. However, existing systems are not ready for automation, or for effective use in continuous operation. Therefore, we have developed theoretical and practical foundations for a new real-time action recognition system. This system is based on Hidden Markov Model (HMM) along with colorizing depth maps. The use of depth cameras provides privacy protection. Colorizing depth images in the hue color space enables compressing and visualizing depth data, and detecting persons. The specific detector used for person detection is You Look Only Once (YOLOv5). Appearance and motion features are extracted from depth map sequences and are represented with a Histogram of Oriented Gradients (HOG). These HOG feature vectors are transformed as the observation sequences and then fed into the HMM. Finally, the Viterbi Algorithm is applied to recognize the sequential actions. This system has been tested on real-world data featuring three participants in a care center. We tried out three combinations of HMM with classification algorithms and found that a fusion with Support Vector Machine (SVM) had the best average results, achieving an accuracy rate (84.04%).

show abstract

Elderly Monitoring and Action Recognition System Using Stereo Depth Camera

Cited by 3 publications

References 2 publications

Real-Time Action Recognition System for Elderly People Using Stereo Depth Camera

Real-Time Action Recognition System for Elderly People Using Stereo Depth Camera

i-Sheet: A Low-Cost Bedsheet Sensor for Remote Diagnosis of Isolated Individuals

HMM-Based Action Recognition System for Elderly Healthcare by Colorizing Depth Map

Contact Info

Product

Resources

About