Vision-based fall detection systems have experienced fast development over the last years. To determine the course of its evolution and help new researchers, the main audience of this paper, a comprehensive revision of all published articles in the main scientific databases regarding this area during the last five years has been made. After a selection process, detailed in the Materials and Methods Section, eighty-one systems were thoroughly reviewed. Their characterization and classification techniques were analyzed and categorized. Their performance data were also studied, and comparisons were made to determine which classifying methods best work in this field. The evolution of artificial vision technology, very positively influenced by the incorporation of artificial neural networks, has allowed fall characterization to become more resistant to noise resultant from illumination phenomena or occlusion. The classification has also taken advantage of these networks, and the field starts using robots to make these systems mobile. However, datasets used to train them lack real-world data, raising doubts about their performances facing real elderly falls. In addition, there is no evidence of strong connections between the elderly and the communities of researchers.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7866979 | PMC |
| http://dx.doi.org/10.3390/s21030947 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Vision-based postural balance assessment of sit-to-stand transitions performed by younger and older adults.
Gait Posture
January 2025
Department of Industrial Engineering and Management, Yuan Ze University, 135 Yuan Tung Road, Chungli District, Taoyuan, Taiwan. Electronic address:
Background: The use of inertial measurement units (IMUs) in assessing fall risk is often limited by subject discomfort and challenges in data interpretation. Additionally, there is a scarcity of research on attitude estimation features. To address these issues, we explored novel features and representation methods in the context of sit-to-stand transitions.
View Article and Find Full Text PDFInterstep Variations of Stairways and Associations of High-Contrast Striping and Fall-Related Events: Observational Study.
Interact J Med Res
January 2025
Department of Kinesiology and Health Science, Sorenson Legacy Foundation Center for Clinical Excellence, Utah State University, Logan, UT, United States.
Background: Interstep variations in step riser height and tread depth within a stairway could negatively impact safe stair negotiation by decreasing step riser height predictability and, consequently, increasing stair users' fall risk. Unfortunately, interstep variations in riser height and depth are common, particularly in older stairways, but its impact may be lessened by highlighting steps' edges using a high-contrast stripe on the top front edge of each step.
Objective: This study aimed to determine (1) if fall-related events are associated with greater interstep riser height and depth variations and (2) if such fall-related events are reduced in the presence of contrast-enhanced step edges compared with a control stairway.
Validity and Reliability of Gait Speed and Knee Flexion Estimated by a Novel Vision-Based Smartphone Application.
Sensors (Basel)
November 2024
Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hong Kong, China.
Patients with knee osteoarthritis walk with reduced speed and knee flexion excursion in the early stance phase. A slow walking speed is also associated with falls in older adults. A novel vision-based smartphone application could potentially facilitate the early detection of knee osteoarthritis and fall prevention.
View Article and Find Full Text PDFReduction of Vision-Based Models for Fall Detection.
Sensors (Basel)
November 2024
Department of Applied Mathematics, University of the Basque Country UPV/EHU, 20600 Eibar, Spain.
Due to the limitations that falls have on humans, early detection of these becomes essential to avoid further damage. In many applications, various technologies are used to acquire accurate information from individuals such as wearable sensors, environmental sensors or cameras, but all of these require high computational resources in many cases, delaying the response of the entire system. The complexity of the models used to process the input data and detect these activities makes them almost impossible to complete on devices with limited resources, which are the ones that could offer an immediate response avoiding unnecessary communications between sensors and centralized computing centers.
View Article and Find Full Text PDFLow-Cost Non-Wearable Fall Detection System Implemented on a Single Board Computer for People in Need of Care.
Sensors (Basel)
August 2024
Centro de Investigación en Mecatrónica y Sistemas Interactivos (MIST), Ingeniería Industrial, Universidad Indoamérica, Av. Machala y Sabanilla, Quito 170103, Ecuador.
Article Synopsis
- A non-wearable fall detection system is proposed using a Raspberry Pi4 and an IR-CUT camera, leveraging deep learning techniques for real-time analysis.
- The system classifies individuals into five states: fallen, crouching, sitting, standing, and lying down, and it sends alerts through Telegram when a fall is detected.
- Key performance metrics include accuracy of 94.8% and sensitivity of 93.1%, and the system prioritizes privacy by not recording video footage unless a fall occurs, offering a promising low-cost solution for fall detection, especially in resource-limited settings.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!