New Advances and Challenges of Fall Detection Systems: A Survey

Xu, Tao; Zhou, Yun; Zhu, Jing

doi:10.3390/app8030418

Cited by 140 publications

(83 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The disadvantage of this method are people need to carry/wear the sensor and the current method can only works properly when the sensor position been defined on the specific position. On the other hand, the model-based approach using machine learning or deep learning to detect more sophisticated pattern than threshold-based detection and it has better accuracy result [3].…”

Section: Related Workmentioning

confidence: 99%

“…In 2017, World Health Organization provide the estimation of the elderly population that has the potential to fall and the results are quite shocking: it is start from 28% until 35% (age > 65 years old) and up to 42% (age >70 years old) [2]. Fall is considered as one of the major health problems in the world and it is on the top two main causes of death [3,4]. If falls occur in this category of people (elderly or people with special health care -for example heart failure), it can cause a major health problem [5,6].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Ans-Assist: Robust Human Fall Detection for Unconstraint Smartphone Positions using Modified Long Short-Term Memory Cell

Astriani*,

Heryadi,

Kusuma

et al. 2019

IJRTE

View full text Add to dashboard Cite

In many aging countries, where the population distribution has shifted to old ages, the need for automatic monitoring devices to help an elderly person when they fall is very crucial. Smartphone is one of the best candidate devices for detecting fall because accelerometer and gyroscope sensors embedded in it respond based on human movements. People usually carry their smartphone in any position and can make fall detection method difficult to detect when fall occurs. This research explored the model for unconstraint human fall detection by using the sensors embedded in smartphone for carried/wearable- sensor-based method. We proposed robust model called Ans-Assist using modified cell of Long Short-Term Memory based model as fall recognition model which can detect human fall from any smartphone position (unconstraint). Some experimental results showed that Ans-Assist achieved 0.95 (± 0.028) average accuracy value using unconstraint smartphone positions. This model can adapt the input from accelerometer and gyroscope sensors which are responsive when human fall.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Ans-Assist: Robust Human Fall Detection for Unconstraint Smartphone Positions using Modified Long Short-Term Memory Cell

Astriani*,

Heryadi,

Kusuma

et al. 2019

IJRTE

View full text Add to dashboard Cite

show abstract

“…But, the development of these systems is very challenging due to light changing, illumination, occlusion, shadows, etc. Many vision-based fall detection methods are proposed [7]. They are based early on extracting the most discriminating hand-crafted features followed by the best classifier such as SVM or KNN.…”

Section: Introductionmentioning

confidence: 99%

Convolutional Neural Network Based on Dynamic Motion and Shape Variations for Elderly Fall Detection

Khraief¹,

Benzarti²,

Amiri³

2019

IJMLC

View full text Add to dashboard Cite

Fall detection became a major concern especially for elderly who lives alone at home. Unexpected situations might happen that influence their health, security and well-being. The development of an intelligent surveillance system is required to alleviate the negative effects of unforeseen circumstances and assisting the elderly in independent living. Currently, convolutional neural network has been successfully used for solving various computer vision tasks, such as object detection and recognition. In this paper, we propose a new vision system for elderly fall detection based on new two stream convolutional neural networks. First, human silhouette is extracted based on background subtraction and person recognition. Second, history of binary motion image HBMI is fed into the first stream characterizing the human shape variations. The second stream is based on amplitude and orientation of optical flow defining the velocity and the direction of the human motion. The system classifies fall events using score fusion schema. Transfer learning is performed to deal with the small amount of fall datasets. Our final network outperforms state-of-the-art results on standard fall datasets.

show abstract

“…Sensing technology can be applied to detect falls, and various systems have been developed [10], [11]. Typically, toilets are installed in a small room and the position of a person using the toilet is also limited.…”

Section: Introductionmentioning

confidence: 99%

Use of Thermal Sensors for Fall Detection in a Simulated Toilet Environment

Shirogane¹,

Takahashi²,

Murata³

et al. 2019

International Journal of New Technology and Research

View full text Add to dashboard Cite

Falls are likely to occur in a toilet. However, people typically are alone when they use the toilet; thus, if a fall occurs, it may be difficult to ask for help. Therefore, we constructed a fall detection system using a thermal image sensor that cannot distinguish details, such as a face, and has high suitability for the toilet environment. In this study, we investigate the influence of the sensor position on the fall discrimination rate. Sensors were installed at four locations and normal toilet movements and falling postures were recorded using eight healthy adults. Discrimination formulas were prepared using thermal images obtained for four subjects, and discrimination rates of falling postures were calculated for the remaining four subjects, which were then compared relative to sensor positions. The highest discrimination rate (95.7%) was obtained with the sensor installed in front of the toilet at the height of the body center of the subject sitting on the toilet. The lowest discrimination rate (60.2%) was obtained with the sensor installed in front of the toilet and close to the ceiling with respect to the subject sitting on the toilet. To detect falls using thermal imaging, sensor positions should be based on various factors, such as whether the entire body can be imaged relative to the falling posture without depending only on the height and direction or whether the difference between normal activity and falling posture is clear, including the environment of the toilet and the assumed fall situation.

show abstract

New Advances and Challenges of Fall Detection Systems: A Survey

Cited by 140 publications

References 31 publications

Ans-Assist: Robust Human Fall Detection for Unconstraint Smartphone Positions using Modified Long Short-Term Memory Cell

Ans-Assist: Robust Human Fall Detection for Unconstraint Smartphone Positions using Modified Long Short-Term Memory Cell

Convolutional Neural Network Based on Dynamic Motion and Shape Variations for Elderly Fall Detection

Use of Thermal Sensors for Fall Detection in a Simulated Toilet Environment

Contact Info

Product

Resources

About