Application of Convolutional Neural Networks for Fall Detection Using Multiple Cameras

Espinosa, Ricardo; Ponce, Hiram; Gutiérrez, Sebastián; Martínez-Villaseñor, Lourdes; Brieva, Jorge; Moya-Albor, Ernesto

doi:10.1007/978-3-030-38748-8_5

Cited by 17 publications

(17 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A machine learning algorithm has been tested. The results exceed those obtained by other authors with the same data set by a significant margin [16,17]. The article is organized as follows.…”

Section: Introductionmentioning

confidence: 63%

“…Therefore, it can be concluded that the original hypothesis is demonstrated, namely that it is possible to detect falls with Model KNN CNN KNN CNN This in [16] in [16] in [17] in [17] the k-nearest neighbor classifier proposed in [16] when using only a single modality (vision from a camera) through the use of human skeleton pose estimate features, obtained via deep neural models, which considerably improves the model performance. Table 3 compares the models with the best performance of camera vision-based systems for fall detection that use the UP-FALL dataset.…”

Section: Fallmentioning

confidence: 96%

“…As an example, in [19] PCA is employed for feature extraction and SVM is employed for classification and in [15] present a VGG-16 architecture modified to receive inputs and CNN for classification. On the other hand, Espinosa et al [17] use a multi-camera approach using UP-FALL dataset, which gets results similar to [19] and [15] using only two cameras. As Table 3 shows, they gets their best performance with a CNN.…”

Section: Fallmentioning

confidence: 99%

See 2 more Smart Citations

Fall Detection using Human Skeleton Features

Ramirez¹,

Velastín²,

Fábregas³

et al. 2021

IET Conf. Proc.

View full text Add to dashboard Cite

A leading cause of death and serious injury in people, especially for the older people, are falls. In addition, fall accidents have a direct economic cost to healthcare systems and have an indirect impact, to the society's productivity. Among the most significant problems in fall detection systems is privacy, limitations of operating devices, and the comparison of machine learning techniques for detection. This article presents a system of fall detection by means of a k-Nearest Neighbor (KNN) classifier based on camera-vision using pose detection of the human skeleton for the features extraction. The proposed method is evaluated with UP-FALL dataset, surpassing the results of other fall detection systems that use the same database. This method achieves a 98.84% accuracy and an F 1 -Score of 97.41%.

show abstract

Section: Introductionmentioning

confidence: 63%

Section: Fallmentioning

confidence: 96%

Section: Fallmentioning

confidence: 99%

See 1 more Smart Citation

Fall Detection using Human Skeleton Features

Ramirez¹,

Velastín²,

Fábregas³

et al. 2021

IET Conf. Proc.

View full text Add to dashboard Cite

show abstract

“…In [18], Ricardo Espinosa used images from the UP-Fall Detection Dataset and 2D CNN model to build a system for fall detection. The author examines images in fixed time window frames, extracting features using an optical flow method that gets relative motion information between two images.…”

Section: Literature Reviews a Related Workmentioning

confidence: 99%

A Feature Selection Approach for Fall Detection Using Various Machine Learning Classifiers

Tran

Dao

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In the study [ 25 , 26 ], the authors have used a multimodal approach, in which the information has been collected from multiple sources such as cameras, microphones, wearable sensors, ambient sensors, smart devices. The combined information helps to improve the classification.…”

Section: Related Workmentioning

confidence: 99%

High Accuracy WiFi-Based Human Activity Classification System with Time-Frequency Diagram CNN Method for Different Places

Sharma

Chao

et al. 2021

Sensors

View full text Add to dashboard Cite

Older people are very likely to fall, which is a significant threat to the health. However, falls are preventable and are not necessarily an inevitable part of aging. Many different fall detection systems have been developed to help people avoid falling. However, traditional systems based on wearable devices or image recognition-based have many disadvantages, such as user-unfriendly, privacy issues. Recently, WiFi-based fall detection systems try to solve the above problems. However, there is a common problem of reduced accuracy. Since the system is trained at the original signal collecting/training place, however, the application is at a different place. The proposed solution only extracts the features of the changed signal, which is caused by a specific human action. To implement this, we used Channel State Information (CSI) to train Convolutional Neural Networks (CNNs) and further classify the action. We have designed a prototype to test the performance of our proposed method. Our simulation results show an average accuracy of same place and different place is 93.2% and 90.3%, respectively.

show abstract

Application of Convolutional Neural Networks for Fall Detection Using Multiple Cameras

Cited by 17 publications

References 58 publications

Fall Detection using Human Skeleton Features

Fall Detection using Human Skeleton Features

A Feature Selection Approach for Fall Detection Using Various Machine Learning Classifiers

High Accuracy WiFi-Based Human Activity Classification System with Time-Frequency Diagram CNN Method for Different Places

Contact Info

Product

Resources

About