Accurate Fall Detection Research Articles

Vision-Based Fall Detection With Multi-Task Hourglass Convolutional Auto-Encoder

Fall detection is a hot research issue in intelligent video surveillance. Falls can generate physical and psychological damage, especially for the elderly. Different from most conventional vision-based fall detection methods typically relying on hand-crafted features, fall detection methods based on deep learning techniques can automatically learn features and hence have got widespread concern recently. However, as deep networks are increasingly applied to fall detection, the problem of information loss in the deep networks can not be ignored, because this will ultimately affect the performance of fall detection. To solve the above problem, we propose a vision-based fall detection method using multi-task hourglass convolutional auto-encoder (HCAE). In this method, hourglass residual units (HRUs) are introduced into the encoder of the HCAE to extract multiscale features by expanding receptive fields of neurons. A multi-task mechanism is presented to enhance the feature representativeness of the network by completing an auxiliary task of frame reconstruction while realizing the main task of fall detection. Experimental results demonstrate that, the proposed method can effectively achieve accurate fall detection with the shallow-layer network, and outperforms several state-of-the-art methods.

Sensor Location Analysis and Minimal Deployment for Fall Detection System

Human falls are considered as an important health problem worldwide. Fall detection systems can alert when a fall occurs reducing the time in which a person obtains medical attention. In this regard, there are different approaches to design fall detection systems, such as wearable sensors, ambient sensors, vision devices, and more recently multimodal approaches. However, these systems depend on the types of devices selected for data acquisition, the location in which these devices are placed, and how fall detection is done. Previously, we have created a multimodal dataset namely UP-Fall Detection and we developed a fall detection system. But the latter cannot be applied on realistic conditions due to a lack of proper selection of minimal sensors. In this work, we propose a methodological analysis to determine the minimal number of sensors required for developing an accurate fall detection system, using the UP-Fall Detection dataset. Specifically, we analyze five wearable sensors and two camera viewpoints separately. After that, we combine them in a feature level to evaluate and select the most suitable single or combined sources of information. From this analysis we found that a wearable sensor at the waist and a lateral viewpoint from a camera exhibits 98.72% of accuracy (intra-subject). At the end, we present a case study on the usage of the analysis results to deploy a minimal-sensor based fall detection system which finally reports 87.56% of accuracy (inter-subject).

Time-Frequency Linearization of Reactive Cortical Responses for the Early Detection of Balance Losses

Aiming at finding a fast and accurate preimpact fall detection (PIFD) strategy, this paper proposes a novel methodology that precociously discriminates the occurrence of unexpected loss of balance from the steady walking, by analyzing the subject’s cortical signal modifications (at the scalp level) in the time-frequency domain. In this study, the subjects were asked to walk at their preferred speed on the treadmill platform programmed to provide unexpected bilateral slippages. The proposed PIFD method exploits synchronously recorded electromyographic (EMG: 2 channels from the same lower limb muscle bundle, bilaterally) and electro-encephalographic (EEG: 13 channels from motor, sensory-motor and parietal cortex areas) signals. To validate the method offline, also, the lower limb kinematics has been reconstructed via a motion capture system (23 reflective markers and 8 fixed cameras). During the PIFD system functioning, the EMG signals from the lateral gastrocnemii are first translated in a binary waveform and then used to trigger the EEG analysis. Once enabled via EMG (every gait cycle), the EEG computation branch extracts and linearizes the rate of variation in the EEG power spectrum density (PSD) for five bands of interests: θ (4–7 Hz), α (8–12 Hz), β I, β II, β III rhythms (13–15 Hz, 16–20 Hz, and 21–28 Hz). The slope of the linearized trend identifies, in this context, the cortical responsiveness parameter. Experimental results from six subjects revealed that the proposed system can distinguish the loss of balance with an overall accuracy of ~96% (average value between sensitivity and specificity). The discrimination process requests, on average, 370.6 ms. This value could be considered suitable for the implementation of countermeasures aimed at restoring the balance of the subject.

Practical Fall Detection System using Vision and Wearable sensors

Fall detection is an important and challenging research problem in healthcare domain. The fall detection system required to operate and give true alert in real time. Many of the existing approaches generates false fall alert which again causes inconvenience for the end users. Hence, there is a need to have robust and accurate fall detection approach with low latency in decision making. In this work, we have proposed and evaluated three different approaches of fall detection system based on a wireless accelerometer based embedded system, RGB Image processing based Software modelling approach and Kinect based depth processing approach. These proposed approaches try to improve on the mentioned drawbacks until we obtain a robust, running in real-time system with high accuracy and low processing time. In all of the demonstrated methods, we do not require any knowledge of the scene and computationally intensive classifiers. The accelerometer based embedded system consists of economic components and is easy to setup. RGB Image processing based Software modelling is simulated on MATLAB have been extensively researched and implemented in real-time. Kinect based depth based techniques are the most recent advancement on the issue and have resolved many discrepancies of the previous methods. The performance of each method is compared against each other. It is shown that our Kinect based depth processing provides promising accuracy of 94% which is better than the other approaches while simultaneously working in real time of 30 frames/second.

A novel fuzzy logic algorithm for accurate fall detection of smart wristband

Falling is a major cause of serious injury or even death for the elderly population. To improve the safety of elderly people, a wide range of wearable fall detection devices have been developed over recent years, such as smart watches, waistbands and other wearable fall detectors. However, most of these fall detection devices are threshold-based and have a high rate of false alarm. This paper presents a novel fuzzy logic fall detection algorithm used in smart wristbands to reduce false alarms and achieve accurate fall detection. Experiments have been conducted in our laboratory and the results show that the proposed algorithm can accurately distinguish fall events from non-fall daily activities such as walking, jumping, clapping, and so forth. It shows good potential for commercial applications.

Development of an Enhanced Threshold-Based Fall Detection System Using Smartphones With Built-In Accelerometers

Falls are a primary accident for elderly people living independently. Obviously, timely and accurate fall detection is critical to reduce the injuries and avoid the loss of life. In order to improve existing smartphone-based fall detection systems, this paper investigates the features of triaxial acceleration values acquired from built-in accelerometers of a smartphone, identifies crucial thresholds of the falls and non-falls, and then proposes an enhanced threshold-based fall detection approach, which could not only distinguish fall events from the most of daily activities (including walking, running, and sitting down), but also support four directions (forward, backward, left lateral, and right lateral) of the falls. In addition, once a falling accident is identified, the user position would be instantaneously transmitted to an emergency center in order to have timely medical assistance. As a consequence, experimental results show the feasibility of our enhanced approach with accuracy and detection rates of 99.38% and 96%, respectively, while a set of 650 test activities including 11 different kinds of daily activities are performed.

Accurate Fall Detection and Localization for Elderly People Based on Neural Network and Energy-Efficient Wireless Sensor Network

Falls are the main source of injury for elderly patients with epilepsy and Parkinson’s disease. Elderly people who carry battery powered health monitoring systems can move unhindered from one place to another according to their activities, thus improving their quality of life. This paper aims to detect when an elderly individual falls and to provide accurate location of the incident while the individual is moving in indoor environments such as in houses, medical health care centers, and hospitals. Fall detection is accurately determined based on a proposed sensor-based fall detection algorithm, whereas the localization of the elderly person is determined based on an artificial neural network (ANN). In addition, the power consumption of the fall detection system (FDS) is minimized based on a data-driven algorithm. Results show that an elderly fall can be detected with accuracy levels of 100% and 92.5% for line-of-sight (LOS) and non-line-of-sight (NLOS) environments, respectively. In addition, elderly indoor localization error is improved with a mean absolute error of 0.0094 and 0.0454 m for LOS and NLOS, respectively, after the application of the ANN optimization technique. Moreover, the battery life of the FDS is improved relative to conventional implementation due to reduced computational effort. The proposed FDS outperforms existing systems in terms of fall detection accuracy, localization errors, and power consumption.

Using Temporal Covariance of Motion and Geometric Features via Boosting for Human Fall Detection.

Fall induced damages are serious incidences for aged as well as young persons. A real-time automatic and accurate fall detection system can play a vital role in timely medication care which will ultimately help to decrease the damages and complications. In this paper, we propose a fast and more accurate real-time system which can detect people falling in videos captured by surveillance cameras. Novel temporal and spatial variance-based features are proposed which comprise the discriminatory motion, geometric orientation and location of the person. These features are used along with ensemble learning strategy of boosting with J48 and Adaboost classifiers. Experiments have been conducted on publicly available standard datasets including Multiple Cameras Fall (with 2 classes and 3 classes) and UR Fall Detection achieving percentage accuracies of 99.2, 99.25 and 99.0, respectively. Comparisons with nine state-of-the-art methods demonstrate the effectiveness of the proposed approach on both datasets.

Accurate Fall Detection in a Top View Privacy Preserving Configuration.

Fall detection is one of the most investigated themes in the research on assistive solutions for aged people. In particular, a false-alarm-free discrimination between falls and non-falls is indispensable, especially to assist elderly people living alone. Current technological solutions designed to monitor several types of activities in indoor environments can guarantee absolute privacy to the people that decide to rely on them. Devices integrating RGB and depth cameras, such as the Microsoft Kinect, can ensure privacy and anonymity, since the depth information is considered to extract only meaningful information from video streams. In this paper, we propose an accurate fall detection method investigating the depth frames of the human body using a single device in a top-view configuration, with the subjects located under the device inside a room. Features extracted from depth frames train a classifier based on a binary support vector machine learning algorithm. The dataset includes 32 falls and 8 activities considered for comparison, for a total of 800 sequences performed by 20 adults. The system showed an accuracy of 98.6% and only one false positive.

Highly Portable, Sensor-Based System for Human Fall Monitoring.

Falls are a very dangerous situation especially among elderly people, because they may lead to fractures, concussion, and other injuries. Without timely rescue, falls may even endanger their lives. The existing optical sensor-based fall monitoring systems have some disadvantages, such as limited monitoring range and inconvenience to carry for users. Furthermore, the fall detection system based only on an accelerometer often mistakenly determines some activities of daily living (ADL) as falls, leading to low accuracy in fall detection. We propose a human fall monitoring system consisting of a highly portable sensor unit including a triaxis accelerometer, a triaxis gyroscope, and a triaxis magnetometer, and a mobile phone. With the data from these sensors, we obtain the acceleration and Euler angle (yaw, pitch, and roll), which represents the orientation of the user’s body. Then, a proposed fall detection algorithm was used to detect falls based on the acceleration and Euler angle. With this monitoring system, we design a series of simulated falls and ADL and conduct the experiment by placing the sensors on the shoulder, waist, and foot of the subjects. Through the experiment, we re-identify the threshold of acceleration for accurate fall detection and verify the best body location to place the sensors by comparing the detection performance on different body segments. We also compared this monitoring system with other similar works and found that better fall detection accuracy and portability can be achieved by our system.

Matching pursuit-based compressive sensing in a wearable biomedical accelerometer fall diagnosis device

There is a significant high fall risk population, where individuals are susceptible to frequent falls and obtaining significant injury, where quick medical response and fall information are critical to providing efficient aid. This article presents an evaluation of compressive sensing techniques in an accelerometer-based intelligent fall detection system modelled on a wearable Shimmer biomedical embedded computing device with Matlab. The presented fall detection system utilises a database of fall and activities of daily living signals evaluated with discrete wavelet transforms and principal component analysis to obtain binary tree classifiers for fall evaluation. 14 test subjects undertook various fall and activities of daily living experiments with a Shimmer device to generate data for principal component analysis-based fall classifiers and evaluate the proposed fall analysis system. The presented system obtains highly accurate fall detection results, demonstrating significant advantages in comparison with the thresholding method presented. Additionally, the presented approach offers advantageous fall diagnostic information. Furthermore, transmitted data accounts for over 80% battery current usage of the Shimmer device, hence it is critical the acceleration data is reduced to increase transmission efficiency and in-turn improve battery usage performance. Various Matching pursuit-based compressive sensing techniques have been utilised to significantly reduce acceleration information required for transmission.

A Mobile Robot for Fall Detection for Elderly-Care

Abstract In 2015, the population of people over the age of 65 is 25.0% in Japan. This means that Japan has already become a super-aging society. In such society, the number of elderly people living alone has been also increased. For such people, a fall accident is serious because it can lead to serious injury or death. Researches and services to monitor behaviours of such people have been proposed. For example, by monitoring the status of use of home appliances, something unusual happened to them can be predicted. However, such systems cannot recognize the detailed behaviours like fall. Surveillance cameras have been introduced only outside the house because of the privacy issues. In this paper, we propose a mobile robot to detect human fall and report it to their observers. The mobile robot consists of a household mobile robot (Yujin Robot's Kobuki), a sensor (Microsoft's Kinect), and a computer (PC) to detect a human and control the robot. For simplicity of the robot and accurate fall detection, the sensor is installed on the robot to follow the target harmoniously. Thus, the sensor can move around with the robot to minimize blind area. The results of our experiments show that improvement of up to 80% in fall detection rate compared to a conventional monitoring technique using position-fixed sensors. Finally, we discuss the capabilities and future works of the robot.

Effective Falls Detection Method Using Two Tri-Axial Accelerometers

Falls detection systems have been developed in recent years because falls are detrimental events that can have a devastating effect on health of the elderly population. Current fall detecting methods mainly employ accelerometer to discriminate falls from activities of daily living (ADL). However, this makes it difficult to distinguish real falls from certain fall-like activities such as jogging and jumping. In this paper, an accurate fall detection system was implemented using two tri-axial accelerometers. By attaching the accelerometers on the chest and the abdomen, our system can effectively differentiate between falls and non-fall events.The Diff_Z and Sum_diff_Z parameter resulted in falls detection rate of 100%, respectively.

Privacy perception and fall detection accuracy for in-home video assistive monitoring with privacy enhancements

Video of in-home activity provides valuable information for assistive monitoring but raises privacy concerns. Raw video can be privacy-enhanced by obscuring the appearance of a person. We consider five privacy enhancements: blur, silhouette, oval, box, and trailing-arrows. We investigate whether a privacy enhancement exists that provides sufficient perceived privacy while enabling accurate fall detection by humans. We recorded 23 1-minute videos involving normal household activities, falling, and lying on the floor after an earlier fall, and created versions of each video for each privacy setting. We conducted an experiment with 376 undergraduate, non-engineering student participants to measure perceived privacy protection and the participant's fall detection accuracy for each privacy setting. Results indicate that the oval provides sufficient perceived privacy for 88% of participants while still supporting fall detection accuracy of 89%, and that the common privacy enhancements blur and silhouette were perceived to provide insufficient privacy.